37857840

Review article

☒ Q Check for updates

Advances in translational research of the rare cancer type adrenocortical carcinoma

Chandrayee Ghosh1, Jiangnan Hu1 & Electron Kebebew1,2 ☒

Abstract

Adrenocortical carcinoma is a rare malignancy with an annual worldwide incidence of 1-2 cases per 1 million and a 5-year survival rate of <60%. Although adrenocortical carcinoma is rare, such rare cancers account for approximately one third of patients diagnosed with cancer annually. In the past decade, there have been considerable advances in understanding the molecular basis of adrenocortical carcinoma. The genetic events associated with adrenocortical carcinoma in adults are distinct from those of paediatric cases, which are often associated with germline or somatic TP53 mutations and have a better prognosis. In adult primary adrenocortical carcinoma, the main somatic genetic alterations occur in genes that encode proteins involved in the WNT-B-catenin pathway, cell cycle and p53 apoptosis pathway, chromatin remodelling and telomere maintenance pathway, CAMP-protein kinase A (PKA) pathway or DNA transcription and RNA translation pathways. Recently, integrated molecular studies of adrenocortical carcinomas, which have characterized somatic mutations and the methylome as well as gene and microRNA expression profiles, have led to a molecular classification of these tumours that can predict prognosis and have helped to identify new therapeutic targets. In this Review, we summarize these recent translational research advances in adrenocortical carcinoma, which it is hoped could lead to improved patient diagnosis, treatment and outcome.

Sections

Introduction

Genomics

Non-coding RNAs

Epigenomics

Steroid hormones and metabolomics

Translational multi-omic analysis

Immune microenvironment

Mouse models

Conclusions

1Department of Surgery, Stanford University, Stanford, CA, USA. 2Stanford Cancer Institute, Stanford University, Stanford, CA, USA. ☒ e-mail: kebebew@stanford.edu

Review article

Introduction

Adrenocortical carcinoma is a rare malignancy with an incidence rate of approximately 1-2 cases per 1 million worldwide1-6. It has a bimodal age distribution with a peak in the first decade of life (5 years of age) and again in the fourth and fifth decades of life4,5,7. The clinical prognosis for adrenocortical carcinoma is poor and depends on patient age, as well as the stage, hormonal secretion status, grade, molecular features (Ki-67 index (a measure of proliferation), types of somatic gene muta- tions, and methylome, microRNA (miRNA) and gene expression profile) and margin status of the tumour4,5,8-10. Paediatric patients present with less-aggressive disease and have a better prognosis with a 5-year over- all survival rate of 54.2-91% (refs. 7,11). In adult patients, the 5-year overall survival rate is 37-59.8% (refs. 1,4,8-10,12,13). Even when adrenocortical carcinoma is localized and a complete resection of the tumour is pos- sible, the 5-year overall survival rate is 75-82% and recurrent disease can occur in one fourth of cases4,10,14,15. Nevertheless, there is great variability in prognosis and the risk of recurrence among patients, who may be classified as having either low-risk disease or high-risk disease based on clinical and pathological prognostic factors. Unfortunately, adrenocortical carcinoma is one of the most treatment-resistant human cancers with currently no curative primary treatment available for those patients with unresectable disease, and moreover, the response rate to the current standard-of-care, first-line treatment (combination chemotherapy with mitotane and etoposide, doxorubicin and cisplatin) for adrenocortical carcinoma is low with only 1.3% of patients achieving a complete response and 19.2% achieving a partial response16,17.

Most adrenocortical carcinomas are sporadic but inherited syn- dromes such as Li-Fraumeni syndrome, Lynch syndrome, multiple endocrine neoplasia type 1 (MEN1), Beckwith-Wiedemann syndrome, familial adenomatous polyposis (FAP), Carney complex, congenital adrenal hyperplasia and neurofibromatosis type 1 (NF1) are associated with a risk of developing adrenocortical carcinoma18-30. Other rare ger- mline variants that potentially predispose individuals to adrenocortical carcinoma have also been described in genes that encode succinate dehydrogenase subunits (SDHx) and armadillo repeat-containing pro- tein 5 (ARMC5)31-33. A substantial number of paediatric adrenocortical carcinomas are associated with a germline TP53 mutation, especially in southern Brazil34-37. There are important differences in the aetiol- ogy and prognosis of adrenocortical carcinoma between paediatric and adult cases and so, where appropriate, we highlight differences between these. However, it should be noted that although this Review encompasses findings from both paediatric and adult adrenocortical carcinomas, most of the current literature on adrenocortical carcinoma is focused on the adult population.

Adrenocortical carcinomas are classified on the basis of their his- tological morphology into the following subtypes: classical subtype (97%), oncocytic subtype (2%) and the rarer subtypes, myxoid subtype (<1%) and sarcomatoid subtype (<1%)38,39. The 2022 WHO classification system also emphasizes the importance of angioinvasion (vascular invasion) in determining the diagnosis and prognosis, and the use of proliferative markers (mitotic count and Ki-67 labelling index) for risk stratification38. In addition to the original widely used and modi- fied scoring systems of Weiss et al.40,41, the latest WHO classification also recognizes the use of other multiparameter diagnostic algorithms (reticulin algorithm, Lin-Weiss-Bisceglia system and Helsinki scoring system) to aid in the diagnosis of adrenocortical carcinomas in adults38. The Helsinki scoring system in particular is most accurate especially when used to assess the less common subtypes of adrenocortical carcinoma39.

There have been substantial advances in our understanding of the molecular basis of adrenocortical carcinoma over the past dec- ade. Thus, in this Review, we focus on the key genetic and epigenetic alterations and metabolomic changes that have been identified in adrenocortical carcinoma, as well as the translational implications of these for improving diagnosis, prognostication and the discovery of new therapeutic targets.

Genomics

In the past decade, comprehensive genomic studies of primary adreno- cortical carcinoma have established dysregulated gene expression and recurrent mutations in genes and their encoded protein signal- ling pathways: zinc and ring finger 3 (ZNRF3; 19-21%) and ß-catenin (CTNNB1;16-20%) in the WNT-B-catenin pathway; p53 (TP53; 16-21%), cyclin-dependent kinase inhibitor 2A (CDKN2A; 15%), RB1 (6.8-7%), cyclin-dependent kinase 4 (CDK4; 2-6.8%) and cyclin E1 (CCNE1; 5.7%) - alterations in TP53, CDKN2A, RB1, CDK4 and CCNE1 collec- tively range from 33% to 44.9% of cases - in the cell cycle and p53 apoptosis pathway; telomerase reverse transcriptase (TERT; 6-14%), telomeric repeat-binding factor 2 (TERF2; 7%), menin (MEN1; 1.5-7%), a-thalassaemia/mental retardation syndrome X-linked (ATRX; 4%) and death domain-associated protein 6 (DAXX; 6%) in the chromatin remod- elling and telomere maintenance pathway; cAMP-dependent protein kinase type la regulatory subunit (PRKARIA; 8-11%) in the cAMP- protein kinase A (PKA) pathway, mediator of RNA polymerase II tran- scription subunit 12 (MED12; 5%) in the DNA transcription pathway, 60S ribosomal protein L22 (RPL22; 7%) in the RNA translation pathway and neurofibromin (NF1; 5%)31,42-44 (Fig. 1). Additional signalling path- ways altered in adrenocortical carcinoma are the fibroblast growth factor (FGF)-FGF receptor (FGFR) and insulin-like growth factor 2 (IGF2)-insulin-like growth factor 1 receptor (IGF1R) pathways45-47. Assie et al.43 performed the first integrated genomic analysis (exome sequencing, transcriptomic and single nucleotide polymorphism (SNP) analysis) of 45 adrenocortical carcinomas. They identified recurrent alterations in known driver genes of other cancer types that were then validated in an independent cohort of 77 adrenocortical carcinomas. In this study, ZNRF3, which encodes an E3 ubiquitin ligase and WNT repressor, was identified as the most frequently altered gene (Fig. 1). As part of the National Cancer Institute (NCI) The Cancer Genome Atlas (TCGA) projects, Zheng et al.44 performed a comprehensive study of the genomic characteristics of 91 primary adrenocortical carcinoma samples from patients across four continents. This study expanded the known adrenocortical carcinoma driver genes to also include PRKAR1A, RPL22, TERF2, CCNE1 and NF1. Genome-wide DNA copy number analysis established that massive DNA loss and whole-genome doubling (WGD) was a frequent occurrence and was associated with advanced disease and poor prognosis. In addition, tumours with increased TERT expres- sion, decreased telomere length and activation of cell cycle programmes were found to be associated with poor prognosis44. Zheng et al.44 also established an adrenocortical differentiation score (ADS) that classified tumours based on the basis of expression of genes associated with adrenocortical differentiation and adrenal function and that was independent of the Weiss histopathology score. In paediatric adreno- cortical carcinoma, the prognostic utility of genomic alterations is less clear as there has been limited research and only small cohorts studied. Although TP53 mutation status was not associated with prognosis in paediatric adrenocortical carcinoma, concomitant mutations in TP53 and ATRX (encodes a catalytic component of the chromatin remodelling complex ATRX-DAXX) have been associated with worse prognosis23,48.

Review article

Fig. 1| Dysregulated molecular pathways in paediatric and adult adrenocortical carcinoma. The pathways most frequently dysregulated (through gene mutations and overexpression or underexpression of genes encoding proteins in these pathways) in adult adrenocarcinoma are p53-RB, WNT, insulin-like growth factor 2 (IGF2), cAMP-dependent protein kinase A (PKA) and telomere maintenance pathways. In paediatric adrenocortical carcinoma, overexpression of MAPK8 and underexpression of PRICKLE, which are WNT- planar cell polarity (PCP) signalling pathway mediators, have also been observed and are associated with poor prognosis but, in adult cases, overexpression of WNT5A and underexpression of PRICKLE were associated with similar outcomes222. In adult adrenocortical carcinoma, constitutive activation of WNT signalling is also found owing to loss-of-function mutations or alterations in the

Paediatric

Adult

WNT pathway

IGF2

WNT

Adenylyl cyclase

IGF1R

FZD

ZNRF3

Tumour cell

ATP

CAMP

PI3K

RAS

PI3K

RAS

1

AKT

RAF

RAC1

APC

AKT

RAF

1

V

PKA regulatory subunit

mTOR

MEK

mTOR

MEK

+

ERK

MAPK8

ß-Catenin

ERK

CDKN2A

Menin

CDKN2A

CDK4

CDK4

PRICKLE

Nucleus

MDM2

Menin

MDM2

Cell cycle

RB

p53

Cell cycle

p53

RB

DAXX

Cyclin E1

ATRX

TERT

Cyclin E1

Upregulated

Downregulated

Altered

DNA repair proteins

tumour suppressor gene encoding adenomatous polyposis coli (APC)43,44, or gain-of-function mutations in the gene encoding ß-catenin, leading to tumorigenesis222. In a whole-genome and transcriptome sequencing study of metastatic adrenocortical carcinomas from seven patients, which integrated the analysis of mutations, RNA expression, mutational signatures and levels of homologous recombination deficiency, it was found that there were alterations in genes involved in DNA repair pathways (MUTYH, BRCA2, ataxia-telangiectasia (ATM), RAD52, mutL homologue 1 (MLH1) and mutS homologue 6 (MSH6)223. ATRX, a-thalassaemia/mental retardation syndrome X-linked; CDK4, cyclin- dependent kinase 4; CDKN2A, cyclin-dependent kinase inhibitor 2A; DAXX, death domain-associated protein 6; FZD, frizzled; IGF1R, insulin-like growth factor 1 receptor; TERT, telomerase reverse transcriptase; ZNRF3, zinc and ring finger 3.

In another recent study of a large cohort (364 cases) of patients with adult adrenocortical carcinoma, additional genomic alterations in several genes such as interleukin 7 receptor (IL7R), low-density lipopro- tein receptor-related protein 1B (LRP1B) and fibroblast growth factor receptor substrate 2 (FRS2) were found in 6%, 8% and 4% of adrenocor- tical carcinomas, respectively49. Mutations in epigenetic regulatory genes were also identified in this study, encompassing the following processes: histone modifications (38%), telomere lengthening (21%), SWI/SNF complex chromatin remodelling (21%) and DNA mismatch repair (MMR) (13.7%). The varied molecular alterations associated with adrenocortical carcinoma prognosis provide insight into why the aggressiveness of the cancer among patients with the same stage of cancer can be so heterogeneous. Furthermore, this comprehen- sive analysis of the genomic changes associated with adrenocortical carcinoma should help to guide future research into which candidate therapeutic agents could be investigated (for example, CDK inhibi- tors, drugs targeting DNA repair pathways and inhibitors of histone acetyltransferases and DNA methylation) and, by extension, stratify patients with adrenocortical carcinomas that could be effectively treated with these candidate targeted agents. Nevertheless, there is

a need to develop in vivo models with such genomic alterations to determine whether they are indeed pathogenic and in turn to test their vulnerability to targeted therapies.

Multi-omic studies (using mRNA, DNA methylation, miRNA and protein analysis) of adrenocortical carcinoma have demonstrated unique subgroups of tumours with distinct biological and clinical features and outcomes43,44. For example, in TCGA study by Zheng et al.44 three subgroups were identified and referred to as Cluster of Cluster (CoC) I, II and III. When these molecular subgroups were compared with other disease types, adrenocortical carcinomas clustered into a distinct homogeneous group separate from adrenocortical adenomas and 10,000 tumours from 33 types of cancer in TCGA database44,50. These findings emphasize that although the molecular features within adrenocortical carcinomas are diverse, this cancer type is unique from others, and future research efforts should focus on the molecular fea- tures specific to adrenocortical carcinoma to further understand the biology of the tumour and to aid the identification of new therapies.

Most adrenocortical carcinomas have at least one alteration in DNA damage repair (DDR) genes, such as those that encode proteins that act as damage sensors (ataxia-telangiectasia mutated (ATM),

Review article

ataxia-telangiectasia and Rad3-related (ATR) and checkpoint kinase 2 (CHEK2)) or are involved in MMR (mutL homologue 1 (MLH1), MLH2, MLH3, mutS homologue 2 (MSH2), MSH3, MSH4, MSH5, MSH6 and PMS2), homologous recombination (p53-binding protein 1 (TP53BP1), BRCA1, BRCA2, BRCA1-interacting protein carboxy-terminal helicase 1 (BRIP1), RAD51 and DNA topoisomerase 3x (TOP3A)), translation synthesis (REV3L; also known as POLZ), base excision repair (BER) (polymerase-ß (POLB)), direct repair (alkB homologue 3 (ALKBH3) and methylguanine-DNA methyltransferase (MGMT)), the Fanconi anae- mia pathway (FANCA and FANCD2) and non-homologous end join- ing (DNA ligase 4 (LIG4), XRCC4 and XRCC6)31,42,51. Many of these gene alterations have been found as both germline and somatic mutations. Furthermore, in TCGA study of Zheng et al.44, identified adrenocor- tical carcinoma driver genes were compared with the Cancer Gene Census52, which revealed that NF1 and MLL4 (also known as KMT2B) were also mutated in more than 5% of the study cohort. Both NF1 and MLL4 are tumour suppressor genes, and mutations in these genes resulting in loss of function or deletion lead to the inability of a cell to perform DDR53-56. Germline MMR gene alterations observed in familial adrenocortical carcinoma cases suggest that these are Lynch syndrome-associated cancers25. Although not all of the DDR gene altera- tions identified in adrenocortical carcinoma or indeed other cancer types have been investigated to determine their pathogenicity, the prevalence of such genetic changes suggests that drugs that target DNA repair pathways should be a focus for future preclinical investigations to ensure that new therapies can be bought to the clinic for patients with adrenocortical carcinoma.

In the study by Zheng et al.44, copy number gains and losses were also observed in up to 61% of adrenocortical carcinomas. This unsta- ble pattern is often associated with WGD, which is a marker of poor prognosis57. Furthermore, this study identified recurrent focal ampli- fications of TERT (5p15.33), TERF2(16q22.1), CDK4 (12q14.1) and CCNE1 (19q12), as well as deletions of RB1 (13q14.2), CDKN2A (9p21.2) and ZNRF3(22q12.1). A focal deletion peak around 4q34.3-4q35.1 was also found, which centred on a long non-coding RNA (lncRNA), LINC00290, previously reported58,59 as being deleted in paediatric adrenocortical carcinomas and other cancer types. Homozygous deletions of ZNRF3 appeared in 16% (n = 14) of tumours evaluated and, with the inclusion of non-silent mutations, the number of adrenocortical carcinomas containing alterations in ZNRF3 increased to 19.3%. These results are in accordance with a previous study that reported recurrent copy number amplification (CNA) at 5p15.33, which includes TERT (14.6%; n=41), and homozygous deletion at 22q12.1, which includes ZNRF3 and another WNT repressor KREMEN1 (9.8% and 7.3% (n = 41), respectively)60.

Overexpression of /GF2 is common in adult and paediatric adren- ocortical carcinoma61-66. IGF2 is expressed from the paternal allele only and, therefore, somatic copy number changes, whereby loss of the maternal allele is accompanied by a duplication of the paternal allele is a likely mechanism to cause IGF2 overexpression61,62,64. The parent-specific expression of IGF2 (known as genomic imprinting) is principally controlled by allele-specific DNA methylation at CTCF-binding sites in the imprinting control region (ICR), located immediately upstream of the neighbouring H19 gene67. Nielsen and colleagues65 analysed gene expression, somatic copy number varia- tion on 11p15.5 (the chromosome on which /GF2 and H19 are located), and DNA methylation status of three differential methylated regions of the IGF2/H19 locus, including the H19 ICR, in adrenocortical carci- noma. IGF2 overexpression was found in 85% of the adrenocortical carcinomas as expected from previous studies61,62,64. The copy number

changes were also associated with hypermethylation of the H19 ICR, implying that the allelic loss was the result of loss of unmethylated maternal alleles. Together, these data provide evidence that loss of the maternal allele is linked to hypermethylation of the H19 ICR and IGF2 overexpression in adrenocortical carcinomas65.

Based on a better understanding of the genomic alterations asso- ciated with adrenocortical carcinomas, the development of future therapeutics will likely be driven by the identification of compounds that effectively target these alterations and selecting such targeted therapy approaches based on the molecular features present in indi- vidual patient tumour samples. As up to 85% of adrenocortical carci- nomas exhibit /GF2 overexpression65 and given that it was shown that inhibiting the IGF2-IGF1R pathway was effective in a xenograft mouse model of adrenocortical carcinoma68, several clinical trials have been conducted using monoclonal antibodies to IGF1R (cixutumumab and figitumumab) as well as a small-molecule inhibitor of IGF1R and the insulin receptor (linsitinib). However, none has resulted in significant clinical efficacy69,70.

Epidermal growth factor receptor (EGFR) is yet another potential therapeutic target in adrenocortical carcinomas as up to 76% (128 of 169 adrenocortical carcinomas had EGFR protein expression) of samples express EGFR71,72. One study in ten heavily pretreated patients with adrenocortical carcinoma using the EGFR inhibitor erlotinib in combination with gemcitabine showed no significant responses73,74. Therefore, despite substantial efforts to improve therapeutic options and molecularly targeted therapies in adrenocortical carcinoma, there has been limited progress. This is mainly due to a limited understanding of the causes of primary therapeutic resistance in adrenocortical car- cinoma. Although tumour cell plasticity has recently been recognized as an important factor in resistance to targeted therapy in other cancer types75,76, its role in adrenocortical carcinoma is unclear. In addition, the heterogeneity of adrenocortical carcinoma, which is reflected in the distinct molecular subtypes that each have unique clinical features and degrees of aggressiveness, suggest that subtyping of the tumour of an individual patient on the basis of genomic profiling to select an appropriate targeted agent could yield improved clinical responses to targeted therapies. Lastly, whole-exome sequencing analyses of metastatic and recurrent adrenocortical carcinomas have shown a higher mutational burden in these tumours compared with primary adrenocortical tumours consistent with tumour heterogeneity77. Thus, genomic profiling of metastatic and recurrent adrenocortical carci- noma samples from a patient to select the appropriate targeted agent is essential but has not been done to date in clinical trials as any altera- tions targeted have been based on analysis of primary adrenocortical carcinomas. Given these considerations, potential additional targeted therapies identified in preclinical studies will be important to move forwards into clinical trials to provide new treatment alternatives for patients with adrenocortical carcinoma78,79 (Table 1).

Non-coding RNAs

miRNAs and lncRNAs regulate post-transcription gene expression and are dysregulated in cancer80,81. The first study to discover an asso- ciation between non-coding RNA dysregulation and adrenocortical carcinoma was with the detection of a lncRNA H19 gene transcript in the 11p15 locus, where /GF2 is also located, and which is associated with Beckwith-Wiedemann syndrome, that results in the development of adrenocortical carcinoma82-84. miRNAs have been identified to be dif- ferentially expressed between benign and malignant adrenocortical tumours and to have a functional role in adrenal tumorigenesis (Fig. 2).

Review article

Table 1| Molecularly targeted agents used in preclinical studies of adrenocortical carcinoma

Category	Agents	Molecular target	Experimental model	Major findings	Refs.
CDK inhibition	Palbociclib (small-molecule inhibitor)	CDK4 and CDK6	In vitro (NCI-H295R ACC cell line and ACC primary cultures)	Concentration-dependent decrease in cell viability	184,185
	Ribociclib (small-molecule inhibitor)	CDK4 and CDK6	In vitro (NCI-H295R and SW-13 ACC cell lines)	Decreased cell viability in SW-13 cells	186
	Flavopiridol (small-molecule inhibitor)	Multiple CDKs	In vitro (NCI-H295R, SW-13 and BD140A ACC cell lines) and in vivo NCI-H295R xenograft model	Combination therapy with a second-generation proteasome inhibitor carfilzomib shows synergistic effect on tumour cell death both in vitro and in vivo	187
IGF1R or mTOR inhibition	NVP-AEW541 (small-molecule inhibitor)	IGF1R	In vitro (NCI-H295R and RL251 ACC cell lines) and in vivo (NCI-H295R and RL251 ACC cell lines) xenograft model	Inhibition of tumour cell proliferation, synergistic inhibition effect with mitotane, a chemotherapy used specifically for adrenocortical carcinoma	68
	Everolimus (RAD-001)	mTOR inhibitor	In vitro (NCI-H295R, SW-13 and HAC15 ACC cell lines and primary ACC tumour culture) and in vivo (NCI-H295R ACC cell lines) xenograft model	Reduces tumour cell growth both in vitro and in vivo	188
WNT-ß-catenin signalling inhibition	PKF115-584 (small-molecule inhibitor)	TCF-ß-catenin complex	In vitro (NCI-H295R ACC cell line)	Inhibits tumour cell growth and induces apoptosis	189
	Doxycycline-induced ß-catenin gene knockdown using shRNA	ß-Catenin	In vitro (NCI-H295R) and in vivo xenograft model	Decreased tumour cell proliferation in vitro; complete absence of tumour growth in vivo in treatment group starting 3 days after tumour cell inoculation	190
PPARY agonism	Rosiglitazone	PPARy as well as other targets	In vitro (NCI-H295R and SW-13 ACC cell lines)	Results in tumour cell growth arrest and cell death; reduces expression of VEGF involved in adrenocortical carcinoma angiogenesis; inhibits the PI3K-AKT and ERK1/2 signalling pathways	191,192
	Pioglitazone	PPARY	In vitro (NCI-H295R ACC cell line)	Inhibition of both the proliferation and invasiveness of NCI-H295R cells in a dose-dependent manner; increased the number of cells in the GO/G1 phase of the cell cycle and decreased the number of cells in S phase; reduced tumour cell invasiveness through Matrigel by approximately 85%	193
Oestrogen pathway inhibition	Hydroxytamoxifen (active metabolite of the oestrogen antagonist tamoxifen)	Oestrogen	In vitro (NCI-H295R ACC cell line) and in vivo xenograft model	Increased expression of the pro- apoptotic factor FASL; reduced tumour cell proliferation in vitro and tumour growth in vivo	194
	G-1 (nonsteroidal GPER agonist)	Oestrogen receptor	In vitro (NCI-H295R cell line) and in vivo xenograft model	Growth inhibitory effect mediated by activation of the ERK1/2 pathway, both in vitro and in vivo	195
	XCT790 (inverse agonist)	ERRa	In vitro (NCI-H295R ACC cell line) and in vivo xenograft model	Reduction of tumour cell growth both in vitro and in vivo; impaired mitochondrial functioning leading to cell death	196
	Progesterone (lipophilic hormone)	Progesterone receptor	In vitro (NCI-H295R, MUC-1 and TVBF-7 cell lines) and in vivo xenograft zebrafish model	Reduction of xenograft tumour area and formation of metastases in embryos; reduction of invasion and induction of apoptosis in vitro	197
Steroidogenesis inhibition	AC-45594 and OOP (inverse agonists of the alkyloxyphenol class and isoquinolinone class, respectively)	SF1	In vitro (NCI-H295R and SW-13 cell lines)	AC-45594 and OOP inhibited tumour cell proliferation of both SF1-positive and SF1-negative cell lines	198

Review article

Table 1 (continued) | Molecularly targeted agents used in preclinical studies of adrenocortical carcinoma

Category	Agents	Molecular target	Experimental model	Major findings	Refs.
Steroidogenesis inhibition (continued)	ATR-101 (small-molecule inhibitor)	ACAT1	In vitro (NCI-H295R ACC cell line) and in vivo xenograft model	Induced tumour cell apoptosis in culture and in xenografts; caused mitochondrial hyperpolarization, release of ROS and ATP depletion; dysregulation of ER calcium stores resulting in ER stress and apoptosis	199,200
	Synthetic HDL nanoparticles (to deplete cholesterol)	SR-BI (which takes up cholesterol from circulating HDL cholesterol)	In vitro (NCI-H295R and SW-13 cell lines)	Increased cell apoptosis induced by mitotane and other chemotherapeutic agents, etoposide or cisplatin	201
Others	Aclarubicin (anthracycline antibiotic)	TOP2A	In vitro (NCI-H295R and SW-13 cell lines)	Reduction in tumour cell proliferation and tumour spheroid size	169
	BI-2536 (small-molecule inhibitor)	PLK1	In vitro (NCI-H295R and SW-13 cell lines) and in vivo xenograft models	Loss of tumour cell viability in vitro of up to 70%; inhibition of growth of SW-13 tumours with less pronounced inhibition of NCI-H295R tumours	202
	Niclosamide (antihelminthic drug)	Parasitic infections (inducing caspase- dependent apopto- sis and G1 cell cycle arrest)	In vitro (NCI-H295R and SW-13 cell lines) and in vivo xenograft model	Inhibition of tumour cell proliferation and migration; decreased tumour growth in vivo with no observed side effects or toxicity	170
	Mebendazole (antihelminthic drug)	Parasitic infections (acts by depolymerizing tubulin)	In vitro (NCI-H295R and SW-13 cell lines) and in vivo xenograft model	Inhibition of tumour cell growth, both in vitro and in vivo; inhibited invasion and migration of tumour cells in vitro and formation of metastases in vivo	203
	Metformin (anti-diabetic drug)	Multiple sites of action	In vitro (NCI-H295R cell line) and in vivo xenograft model	Inhibition of tumour cell proliferation in vitro and reduced tumour growth in vivo	204
	miRNA-7 (small non-coding RNA systemically delivered in nanoparticles)	miRNA-7 targets	ACC cell lines (NCI-H295R and SW-13) and in vivo xenograft model	Reduced tumour cell growth in vitro and in vivo	205

ACAT1, acyl-coenzyme A:cholesterol O-acyltransferase 1; ACC, adrenocortical carcinoma; CDK, cyclin-dependent kinase; ER, endoplasmic reticulum; ERRa, oestrogen-related receptor-a; FASL, FAS ligand; GPER, G protein-coupled oestrogen receptor; HDL, high-density lipoprotein; IGF1R, insulin-like growth factor 1 receptor; PLK1, polo-like kinase 1; PPAR, peroxisome proliferator-activated receptor; ROS, reactive oxygen species; SF1, steroidogenic factor 1; shRNA, short hairpin RNA; SR-BI, scavenger receptor class B type I; TCF, T cell factor; TOP2A, DNA topoisomerase 2a; VEGF, vascular endothelial growth factor.

Several studies have compared miRNA expression in adrenocortical carcinomas with normal adrenal cortex and adrenocortical adeno- mas. One study showed that miR-483-5p and miR-195 are differentially expressed in adrenocortical carcinoma, with high miR-483-5p and low miR-195 expression being associated with aggressive clinicopathologi- cal features and poor prognosis85. In another study of 46 primary and 2 recurrent adrenocortical carcinoma tumour samples, a dozen miRNAs were found to be upregulated (for example, miR-503) or downregulated (for example, miR-34a and miR-497) in carcinomas compared with adenomas86. Upregulation of miR-503 in adrenocortical carcinoma was the best single discriminator between carcinoma and adenoma and the combination of miR-34a and miR-497 underexpression discriminated carcinomas from adenomas with 100% sensitivity and 96% specificity; there was also found to be good correlation between the Weiss scoring system and miRNA expression levels86. Unlike prior studies that used microarray and quantitative PCR with reverse transcription (RT-qPCR) techniques, recent studies using next-generation sequencing (NGS) of RNA have identified additional differentially expressed miRNAs, such as miR-503-5p, as well as detecting 411 expressed miRNAs that exist in 1,763 various length isoforms in 51 unique sets of samples of adren- ocortical carcinoma, adrenocortical adenoma and normal adrenal

cortex87. Moreover, 15 miRNAs differentiated between adrenocortical carcinoma and non-malignant (adrenocortical adenomas and normal adrenal cortex) samples. Of these, expression levels of six miRNAs (miR- 503-5p, miR-483-3p, miR-450a-5p, miR-210, miR-483-5p and miR-421) distinguished adrenocortical carcinoma from non-malignant samples with 95% accuracy with the best single miRNA malignancy marker being miR-483-3p (Table 2).

miRNAs exist in protein complexes or within extracellular vesicles shed from tumour cells, where they are relatively stable and protected from enzymatic degradation in the circulation88,89. Thus, they have been evaluated as potential noninvasive diagnostic markers for various cancer types. Serum miR-34a and miR-483-5p levels have been reported to distinguish between aggressive and non-aggressive adrenocorti- cal carcinoma and between benign and malignant adrenocortical tumours90-92 (Table 2). Furthermore, both miRNAs were found to be secreted from adrenocortical carcinoma cells using in vitro culture systems92. Upregulated miRNAs in adrenocortical carcinoma can regulate various targets such as LIN28 (which encodes a suppressor of miRNA biogenesis), programmed cell death protein 4 (PDCD4) and PUMA (which encodes a mediator of apoptosis), whereas downregu- lated miRNAs regulate targets such as RAF, EGFR, p21-activated kinase 1

Review article

(PAK1), cyclin-dependent kinases regulatory subunit 2 (CKS2), CDK1, IGFR1, MTOR, TARBP2 (which encodes a protein required for formation of the RNA-induced silencing complex), DICER1 (which encodes an endoribonuclease involved in the maturation of miRNAs), argonaute 2 (AGO2; encodes a protein required for RNA-mediated gene silencing), ZNF367,BCL2, metadherin (MTDH; also known as LYRIC), ZEB1, MALAT1, eukaryotic translation initiation factor 4E (EIF4E) and splicing factor, proline- and glutamine-rich (SFPQ; encodes a pre-mRNA splicing factor)88,93-97. In an integrated analysis of differentially expressed genes and differentially expressed miRNAs that target these differentially expressed genes using samples of adrenocortical carcinoma, adreno- cortical adenoma and normal adrenal cortex, several key molecu- lar pathways were identified to be enriched: protein ubiquitylation, andnitricoxide, PI3K-AKT,arylhydrocarbonreceptor, CDC42,mTORand oncostatin M signalling98. Implications of these dysregulated molecular pathways in adrenocortical carcinoma were then evaluated with in vitro studies using two adrenocortical carcinoma cell lines. As an example, it was found that oncostatin M inhibited adrenocortical carcinoma cellular proliferation by inducing apoptosis through increasing cas- pase 3 and caspase 7 enzyme activity and promoting the cleavage of full-length poly[ADP-ribose] polymerase (PARP) and caspase 3 in vitro.

lncRNAs are differentially expressed in adrenocortical carcinoma compared with normal adrenal cortex and benign adrenocortical tumours, and may have diagnostic and prognostic utility, as well as being potential therapeutic targets99,100. However, there have been a limited number of lncRNA profiling studies in adrenocortical car- cinoma tissue samples. Low expression of PRINS and LINC00271 is associated with adrenocortical carcinoma recurrence and metastasis, and poor prognosis, respectively99,100. Another lncRNA downregulated

in adrenocortical carcinoma is ASB16-AS1, which is also associated with poor prognosis101. This lncRNA functions as a tumour suppressor in adrenocortical carcinoma by regulating the expression of IGF1R and CDK6 (ref. 101). UCA1 is a lncRNA that is overexpressed and associ- ated with poor prognosis in adrenocortical carcinoma; functional studies demonstrate that UCA1 regulates tumour cell proliferation by reducing the abundance of miR-298, which in turn increases CDK6 expression as miR-298 binds to the 3’ untranslated region (UTR) of CDK6 (ref. 102). Overall, studies profiling the expression of miRNAs and lncRNAs in adrenocortical carcinoma show that they can be used to distinguish between benign and malignant tumours, and are associated with prognosis. The data for the role of miRNAs in adreno- cortical carcinoma are more robust with similar alterations found across multiple studies. For example, miR-210 (refs. 87,90,103,104), miR-483-5p43,44,85,87,90,105,106 and miR-503 (refs. 43,85,87,90) are upregu- lated and miR-195 (refs. 43,85,90,104-106) is downregulated in adreno- cortical carcinoma compared with benign and normal adrenocortical tissue samples. Additionally, the possibility of detecting altered miR- NAs and lncRNAs in liquid biopsy samples makes their likely transla- tional impact more immediate for the diagnosis and prognostication of adrenocortical carcinoma.

Epigenomics

Epigenetic changes, such as changes in DNA methylation and histone modifications, have been implicated in adrenocortical carcinoma107. A genome-wide DNA methylation profiling study in adrenocortical tumours showed that CpG methylation patterns are distinctly different between normal, benign, primary malignant and metastatic adreno- cortical tissue samples108. Specifically, adrenocortical carcinomas

Fig. 2 | Dysregulated molecular pathways owing to epigenomic changes in adult adrenocortical carcinoma. Genes that encode proteins that regulate two important pathways (WNT-ß catenin and insulin-like growth factor 2 (IGF2) signalling) involved in adrenocortical carcinoma are functionally altered by epigenetic changes, namely promoter region hypomethylation and hypermethylation. One study using an original patient data set reported higher methylation of PR domain zinc finger protein 5 (PRDM5) and Dickkopf-related protein 3 (DKK3), which both encode inhibitors of WNT signalling, resulting in lower mRNA expression, and hypomethylation of T-box protein 3 (TBX3), which is a WNT target gene, resulting in overexpression of TBX3 mRNA224. These correlations were also observed in The Cancer Genome Atlas (TCGA) data set224. N-MYC downstream-regulated gene 1 (NDRG1) mRNA and mRNA coding for related family members were also found to be underexpressed in adrenocortical

IGF2

Fly

WNT

IGF1R

FZD

W

Tumour cell

Nucleus

Hypermethylated

GATA4

NDRG1

CDKN2A

DLEC1

H19

PRDM5

DKK3

HDAC10

PYCARD

SCGB3A1

Hypomethylated

CTNBB1

TBX3

Upregulated

Downregulated

carcinoma compared with normal adrenal cortex, owing to hypermethylation of the genes224. In addition to mutations in the gene that encodes ß-catenin, CTNBB1 hypomethylation has also been reported in some studies107. Another comprehensive study that compared adrenal tumours (benign and malignant) with normal adrenal cortex showed hypermethylation of the genes GATA4, deleted in lung and oesophageal cancer protein 1 (DLEC1), cyclin-dependent kinase inhibitor 2A (CDKN2A), histone deacetylase 10 (HDAC10), PYD and CARD domain-containing protein (PYCARD) and secretoglobin family 3A member 1 (SCGB3A1; also known as HIN1), resulting in lower mRNA expression in the adrenal tumours compared with the normal adrenal cortex225. H19 promoter hypermethylation has been shown to cause the altered expression of both H19 and IGF2 in adrenocortical carcinoma in multiple studies109,110,226. FZD, frizzled; IGF1R, insulin-like growth factor 1 receptor.

Review article

Table 2 | Select diagnostic markers in adult and paediatric patients with adrenocortical carcinoma

Diagnostic marker	Sample	Patient type	Method of detection	Type of diagnosis	Refs.
SF11	Tissue	Adult and paediatric	IHC	Disease confirmation	206
Ki-67 1	Tissue	Adult and paediatric	IHC	Diagnosis of uncertain ACC cases and metastasis	206
IGF2 1	Tissue	Adult	IHC	Disease confirmation	206
ADIPOR1 and ADIPOR2 1	Tissue	Adult	IHC	Disease confirmation	206
ß-Catenin 1	Tissue	Adult and paediatric	IHC	Disease confirmation	207
IGF1R 1	Tissue	Adult and paediatric	IHC	Disease confirmation	206
MYC 1	Tissue	Adult	IHC	Primary ACC identification	208
IL13RA2 1	Tissue	Adult	Genome-wide gene expression profile by RT-qPCR	Distinguishing benign from malignant adrenocortical tumours	209
HTR2B 1	Tissue	Adult	Genome-wide gene expression profile by RT-qPCR	Distinguishing benign from malignant adrenocortical tumours	209
CCNB2 ↑	Tissue	Adult	Genome-wide gene expression profile by RT-qPCR	Distinguishing benign from malignant adrenocortical tumours	209
RARRES2 1	Tissue	Adult	Genome-wide gene expression profile by RT-qPCR	Distinguishing benign from malignant adrenocortical tumours	209
SLC16A9 1	Tissue	Adult	Genome-wide gene expression profile by RT-qPCR	Distinguishing benign from malignant adrenocortical tumours	209
MCM3 and MCM7 1	Tissue	Adult	IHC	Distinguishing benign from malignant adrenocortical tumours	210
miR-483-3p 1	Tissue and blood	Adult	NGS, RT-qPCR	Disease confirmation	87
Methylation 1 of regulatory regions of IGF2 and H19 Expression 1 of IGF2 and H19	Tissue	Adult	MS-MLPA, RT-qPCR	Biomarkers, ACC confirmation	109,111

The most widely studied and most significantly differentially expressed markers are listed. ACC, adrenocortical carcinoma; ADIPOR, adiponectin receptor; CCNB2, cyclin B2; HTR2B, 5-hydroxytryptamine receptor 2B; IGF2, insulin-like growth factor 2; IGF1R, insulin-like growth factor 1 receptor; IHC, immunohistochemistry; IL13RA2, interleukin 13 receptor subunit &2; MCM, minichromosome maintenance protein; MS-MLPA, methylation-specific multiplex ligation-dependent probe amplification; NGS, next-generation sequencing; RARRES2, retinoic acid receptor responder 2; RT-qPCR, quantitative PCR with reverse transcription; SF1, steroidogenic factor 1; SLC16A9, solute carrier family 16 member 9.

have globally hypomethylated DNA compared with normal and benign adrenocortical tissue. The methylation status of the promoter regions associated with H19 and /GF2 have also been reported to be diagnostic markers for adrenocortical carcinoma107,109. Barreau et al.110 observed that CpG methylation status clustered adrenocortical carcinoma into two distinct groups. One group of adrenocortical carcinomas were slightly hypermethylated compared with adrenocortical adenomas, and the other group was hypermethylated compared with not only adrenocortical adenomas but also adrenocortical carcinomas of the first group110. The latter group had two subgroups: a CpG island methylator phenotype (CIMP)-high subgroup and a CIMP-low sub- group. The level of CpG methylation was also associated with sur- vival; adrenocortical carcinomas with CIMP had a worse prognosis than non-CIMP tumours110. Jouinot et al.111 conducted a study in which they used methylation-specific multiplex ligation-dependent probe amplification (MS-MLPA), a form of targeted methylation assay, to measure methylation in samples from patients with adrenocorti- cal carcinoma and to identify a set of probes that could be used to generate a prognostic marker. They identified that the mean meth- ylation of four genes (paired box 5 (PAX5), glutathione S-transferase P (GSTP1), PYD and CARD domain-containing (PYCARD) and PAX6) was the strongest methylation marker, and that the four-gene methyla- tion status was an accurate prognostic factor for disease-free survival and overall survival.

In TCGA adrenocortical carcinoma study44, three methylation subtypes were observed (CIMP low, intermediate and high), which in turn enabled the cohort to be classified into three survival groups with 92.4% accuracy that were validated in an independent cohort. In a targeted pyrosequencing analysis of the promoter regions of five pre- selected genes (PAX5, GSTP1, PYCARD, PAX6 and GO/G1 switch protein 2 (GOS2)), Lippert and colleagues112 found that the methylation status of PAX5 was an independent prognostic factor and that including this marker along with the tumour stage, grade and margin status, and patient age and symptoms was most accurate for predicting prognosis in 237 patients with adrenocortical carcinoma. In another targeted methylation analysis, hypermethylation and silencing of GOS2 was found to be exclusively present in the CIMP-high group and was an independent predictor of shorter disease-free survival and overall survival in patients with adrenocortical carcinoma113. Finally, the IGF2 methylation score (based on the methylation status of three regula- tory regions of IGF2) has been reported to have high diagnostic accu- racy for adrenocortical carcinoma and to predict the development of metastases on univariate analysis.

Clay and colleagues114 conducted a DNA methylation analysis of 48 adrenocortical tumour samples from the International Paediat- ric Adrenocortical Tumour Registry along with 12 paediatric normal adrenal cortex controls and compared their results with the adult adrenocortical carcinoma cohort from TCGA study44. They identified

Review article

two distinct paediatric adrenocortical tumour methylation groups with differences in clinicopathological features and patient outcome. One group was enriched for CTNNB1 variants and had more-aggressive tumours. The other group was enriched for TP53 germline variants, had less-aggressive tumours and a younger age at presentation. The CIMP groups identified in TCGA cohort44 were not present in the pae- diatric adrenocortical carcinoma samples. Furthermore, integration of the methylome features of paediatric adrenocortical carcinomas with histopathological features using the Wieneke criteria identified a high-risk group with a fatal disease course. Thus, together, these studies suggest that DNA methylation analysis can enhance prog- nostication in adrenocortical carcinoma and that the methylome of paediatric adrenocortical carcinomas is different from that of adult adrenocortical carcinomas. Furthermore, the analysis of select gene promoter regions using clinically available samples can be used to better predict recurrence-free survival and overall survival in patients with adrenocortical carcinoma, as well as to inform clinical decisions on the need for established adjuvant treatment.

Screening for histone methyltransferases, demethylases and associated factors in publicly available data of adrenocortical car- cinoma showed that the histone methyltransferase enzyme EZH2 is overexpressed115. A high level of expression of EZH2, which results from a deregulated p53-RB-E2F pathway, was associated with increased adrenocortical carcinoma cell proliferation in vitro as well as poor prognosis in patients with adrenocortical carcinoma115. EZH2 also cooperates with the transcriptional activator E2F1 to stimulate expression of genes involved in adrenocortical carcinoma progres- sion, and co-overexpression of EZH2 and E2F1 is also associated with poor prognosis116. Zheng et al.44 also reported that 22% of the analysed adrenocortical carcinoma samples in their study had dysregulated expression levels of mRNAs encoding proteins involved in histone modifications (MLL, MLL2 and MLL4) and chromatin remodelling (ATRX and DAXX). Interestingly, 7% of adrenocortical carcinoma cases have a mutation in the MEN1 gene, which encodes the tumour suppressor menin43 (Fig. 1), which has been reported to interact with the histone methyltransferases MLL and MLL2.

These findings taken together show not only that analyses of epig- enomic changes in adrenocortical carcinoma can refine diagnosis and prognostication but that treatment strategies that target these spe- cific epigenomic alterations (for example, histone methyltransferase inhibitors) could represent a new therapeutic approach.

Steroid hormones and metabolomics

Both adrenocortical carcinoma and adrenocortical adenoma originate from steroidogenic cells in the adrenal cortex and can overproduce steroid hormones, leading to clinical syndromes such as Cushing syndrome, virilization or primary hyperaldosteronism117,118. The adult adrenal cortex produces mainly three classes of steroid hormone, which are mineralocorticoids, glucocorticoids and androgens (Fig. 3). Identi- fying key regulators of steroidogenesis in adrenocortical carcinoma is an important focus of current research, as diagnostic and prognostic markers as well as therapeutic targets might be discovered.

Currently, tumour size, growth and imaging characteristics are used to differentiate benign adrenocortical adenomas and other adre- nal masses from adrenocortical carcinoma but these are imperfect. As a result, steroid profiling has emerged of late as an effective, nonin- vasive diagnostic and prognostic test for adrenocortical carcinoma. Taylor et al.119 suggested that analysis of a serum steroid panel may help in the differential diagnosis of adrenocortical carcinomas and

adrenocortical adenomas and so evaluated the diagnostic potential of a 13-steroid plasma panel in a large series of patients with these tumours using combined liquid chromatography-tandem mass spectrometry (LC-MS/MS) and machine learning techniques119. From this method- ology, they were able to identify a reliable diagnostic signature for adrenocortical carcinoma (Table 3). In another study of a cohort of 135 postoperative patients, usinggas chromatography-mass spectrome- try (GC-MS)-based steroid profiling and machine learning in a form of urine steroid metabolomics, revealed a urine ‘steroid fingerprint’ at recurrence, which resembled that observed before complete resection in most cases120. Moreover, recurrence detection by this steroid pro- filing method preceded detection achieved by imaging by more than 2 months in some patients. Thus, urine steroid metabolomics looks as though it might have potential as a clinical tool for the detection of adrenocortical carcinoma recurrence120. However, there are currently no established standards for quantifying steroid levels by LC-MS/MS or GC-MS and normal reference ranges, and few clinical laboratories have such equipment available for routine testing. Furthermore, LC- MS/MS and GC-MS are currently costly, and samples are not rapidly processed. Yet, unfortunately, practical methods, aside from LC-MS/ MS or GC-MS, to detect or predict adrenocortical carcinoma behav- iour and prognosis are constrained by their lack of clinical availability. Thus, to overcome the aforementioned limitations of LC-MS/MS and GC-MS, an immunoassay was developed to measure levels of 12 serum steroid metabolites in adrenocortical carcinomas and adenomas, both of which can produce cortisol, which complicates diagnosis121. Suzuki and colleagues121 found that basal levels of steroid precursors were significantly higher in adrenocortical carcinoma than in adenoma; in particular, 17-hydroxypregnenolone and 11-deoxycorticosterone had the highest sensitivity and specificity based on the area under the receiver operating characteristic (ROC) curve (AUC) analyses in distinguishing between adrenocortical carcinoma and adenoma (Fig. 3 and Table 3). Androstenedione and dehydroepiandrosterone sulfate (DHEAS) levels in combination also had high specificity with high accuracy in distinguishing between adrenocortical carcinoma and adenoma. Furthermore, serum levels of 11-deoxycortisol were signifi- cantly associated with poor prognosis based on the European Network for the Study of Adrenal Tumours (ENSAT) adrenocortical carcinoma classification (a revised tumour-node-metastasis (TNM) staging system of adrenocortical carcinoma), and levels of testosterone were significantly associated with the tumour Ki-67 index in both men and women. Overall, this study highlights how combined measurements of serum steroid metabolites can be a useful noninvasive method for the diagnosis of adrenocortical carcinoma as well as providing additional prognostic information when used in conjunction with clinical and pathological information121.

Steroid profiling of adrenocortical carcinoma with LC-MS/MS can also be combined with patient clinical data to give clinical diag- nostic utility121,122. In a prospective multicentre study with adult parti- cipants who had newly diagnosed adrenal masses, the accuracy of diagnostic imaging strategies based on maximum tumour diameter (≥4 cm compared with <4 cm), imaging characteristics (positive versus negative) and urine steroid metabolomics (low, medium or high risk of adrenocortical carcinoma) were assessed individually as well as in combination. The investigators observed that the three tests in combi- nation improved the accuracy of detection of adrenocortical carcinoma tumours compared with tumour size and imaging characteristics (currently used for risk stratification for malignancy) and urine steroid metabolomics alone122.

Review article

Metabolomic profiling in patients with adrenocortical adenoma or carcinoma have shown that the urinary metabolome is different between these two patient groups and that some metabolites are likely directly secreted from the tumours123. Patel et al.123 performed an untargeted metabolomic analysis using ultra-performance liquid chromatography-mass spectrometry (UPLC-MS) to determine a novel and specific urinary metabolomic signature that could discriminate patients with benign adrenal tumours from those with adrenocortical carcinoma. In this study, 69 unique metabolites were discovered, and four known metabolites were identified. Levels of urinary creatine ribo- side were increased in patients with adrenocortical carcinoma, whereas levels of L-tryptophan, NE,NE,NE-trimethyl-L-lysine and 3-methylhisti- dine were lower in patients with adrenocortical carcinoma. Combined multivariate analysis of the four biomarkers showed an AUC of 0.89 (sensitivity of 94.7%, specificity of 82.6%) for distinguishing adreno- cortical carcinoma from benign tumours. However, of these four bio- markers identified in the training set, only creatine riboside could be validated, along with four other unknown metabolites. Furthermore, direct analysis of adrenocortical tumour samples showed that creatine riboside, NE,NE,NE-trimethyl-L-lysine and two of the unknown metabo- lites were elevated in these tissues, suggesting that their levels were elevated as a result of secretion directly from the tumour. Although the function of creatine riboside has yet to be revealed, the elevated levels in adrenocortical carcinoma may indicate that this metabolite

is involved in tumorigenesis123. Elevated urinary creatine riboside has also been found to be associated with poor prognosis in patients with lung cancer124, and was a diagnostic and prognostic marker in patients with intrahepatic cholangiocarcinoma125, suggesting that it may be a universal urinary marker of cancer.

An important mechanism controlling steroidogenesis and the activity of steroid hormones in tissues is sulfation and desulfation of steroids by ubiquitously expressed steroid sulfate enzymes126. In a recent study by Sun and colleagues127, matrix-assisted laser desorption/ ionization (MALDI) mass spectrometry imaging (MALDI-MSI) was used to analyse steroid sulfation in 72 adrenocortical carcinoma tissue sam- ples. Low levels of oestradiol-17§ 3-sulfate and oestrone 3-sulfate and the presence of the [M-H-H2O] adduct of oestradiol-17฿ 3,17-disulfate in adrenocortical carcinoma tissue samples were associated with worse prognosis. Furthermore, analysis of the expression of enzymes associ- ated with sulfation of steroid hormones (sulfotransferases (SULTs), which catalyse sulfation and steryl-sulfatase (STS), which catalyses desulfation of steroid hormones) by immunohistochemistry showed a significant association with overall survival (higher levels of SULT2A1, SULT1E1 and STS were associated with better prognosis). Both the ster- oid sulfation analysis by MALDI-MSI and the immunohistochemistry for expression of the enzymes were performed in tissue microarrays generated from adrenocortical carcinoma tissue paraffin blocks. The use of already available clinical tissue samples for this analysis suggests

Fig. 3 | Steroidogenesis pathway of the adrenal cortex and its dysregulation in adrenocortical carcinoma. Here, we have classified steroids and their metabolic byproducts from enzymatic reactions as overproduced based on the reports of recent studies that detected them in serum and/or urine and found their levels increased in adrenocortical carcinomas compared with adrenocortical adenomas121,227. For simplicity, we only highlight the enzymes

Adrenal gland

Medulla (ectodermal origin)

Cortex (mesodermal origin)

Mineralocorticoids Cholesterol

Glucocorticoids

Sex hormones

5-PT

DHEAS

6a-Hydroxy-DHEA

StAR and p450scc

SULT2A1

STS

5-PD

Pregnenolone

3ß-HSD

17a-Hydroxylase

17-Hydroxypregnenolone

3ß-HSD

17,20-Lyase

DHEA

17ß-HSD

Androstenediol

3ß-HSD

3ß-HSD

PD

Etio

Progesterone

17a-Hydroxylase

17-Hydroxyprogesterone

17,20-Lyase

Androstenedione

17ß-HSD

Testosterone

21-Hydroxylase

21-Hydroxy

Aromatase

Aromatase

THDOC 5a-THDOC

11-Deoxycorticosterone

11-Deoxycortisol

TSH

Androsterone

11ß-Hydroxylase

11ß-Hydroxylase

Oestrone

17ß-HSD

Oestradiol

Corticosterone

Cortisol

60-OHF

18-Hydroxylase

18-Hydroxycorticosterone

11-Oxidase

Overproduced steroids

Aldosterone

Overproduced byproducts of steroid metabolism

that metabolize the steroid hormones in the various pathways. 5-PD, 45-pregnenediol; 5-PT, 45-pregnenetriol; 6ß-OHF; 6ß-hydroxycortisol; DHEA, dehydroepiandrosterone; DHEAS, dehydroepiandrosterone sulfate; Etio, etiocholanolone; HSD, hydroxysteroid dehydrogenase; STS, steryl-sulfatase; SULT2A1, sulfotransferase 2A1; THDOC, tetrahydrodeoxycorticosterone; TSH, thyroid-stimulating hormone.

Review article

Table 3 | Select prognostic markers in adult and paediatric patients with adrenocortical carcinoma

Prognostic marker class	Specific markers under each class	Sample	Patient type	Method of detection	Prognostic value	Refs.
miRNAs	miR-483-5p, miR-139-5p, miR-34a, miR-100, miR-181b, miR-184, miR-210, miR-101, miR-320b, miR-27a-3p, miR-22-3p, miR-210-3p 1 miR-335, miR-195, miR-376a V	Tissue and blood	Adult	RT-qPCR	OS	211,212
IncRNA	PRINS 1 RAD50 1 HAND2 V	Tissue	Adult	RT-qPCR RT-qPCR RT-qPCR, microarray	RFS RFS OS	99,100
Urine steroids	THS 1 5-PD 1	Urine	Adult	LC-MS/MS	RFS; postoperative recurrence	122
	PT 1 5-PT 1
	THDOCT
	Etio 1
	5a-THA 1
CTCs	Single CTC 1	Blood	Adult	CTC isolation, staining and identification and characterization	OS and mortality	213
	CTC clusters 1				indicator in pre-surgery	213
	CAMLs 1					213
Cell-free circulating tumour DNA	Quantity 1	Blood	Adult	NGS from blood sample	Disease progres- sion in specific subgroups of ACC	214
Pathology markers	Ki-67 ↑	Tissue	Adult and paediatric	IHC; correlations analysed using chi-square tests and Pearson's or Spearman's tests	Proliferation index for OS and RFS	215
Molecular markers	ZWINT 1	Tissue	Adult	Enrichment analysis from microarray; statistical analysis	Disease progression	216
	PRC11
	CDKN3 1
	CDK11
	CCNA2 1
	BUB1B-PINK1 1	Tissue	Adult	qPCR, statistical analysis	OS	217
	DLGAP5-PINK11	Tissue	Adult		Disease-free survival	217
	VAV2 1	Tissue	Adult	IHC	RFS; OS	218
	IGF2 1	Tissue	Adult	IHC	OS	219
	IGF1R 1	Tissue	Paediatric	IHC	OS and metastasis	219
	SF1 1	Tissue	Adult and paediatric	IHC	RFS; OS	220
	ß-catenin 1	Tissue	Adult and paediatric	IHC	OS	177
Mitotic count		Tissue	Adult	Microscopy	RFS; OS	221
Methylation markers (CIMP or non-CIMP)	PAX5 1 me	Tissue	Adult	MS-MLPA	RFS; OS	111
	PAX6 1 me
	PYCARD 1 me
	GSTP1 1 me

The most widely studied and most significantly differentially expressed markers are listed. 5a-THA, 5a-tetrahydro-11-dehydrocorticosterone; 5-PD, 45-pregnenediol; 5-PT, 45-pregnenetriol; ACC, adrenocortical carcinoma; BUB1B, BUB1 mitotic checkpoint serine/threonine kinase B; CAML, cancer-associated macrophage-like cell; CCNA2, cyclin A2; CDK1, cyclin-dependent kinase 1; CDKN, cyclin-dependent kinase inhibitor; CDKN3, cyclin-dependent kinase inhibitor 3; CIMP, CpG island methylator phenotype; CTC, circulating tumour cell; CTNNB1, B-catenin; Etio, etiocholanolone; GSTP1, glutathione S-transferase P; HDAC10, histone deacetylase 10; IGF2, insulin-like growth factor 2; IGF1R, insulin-like growth factor 1 receptor; IHC, immunohistochemistry; LC-MS/MS, liquid chromatography-tandem mass spectrometry; IncRNA, long non-coding RNA; me, methylation; miR, microRNA; MS-MLPA, methylation-specific multiplex ligation-dependent probe amplification; OS, overall survival; PAX, paired box; PINK1, PTEN-induced kinase 1; PRC1, protein regulator of cytokinesis 1; PRINS, psoriasis-associated non-protein coding RNA induced by stress; PT, pregnanetriol; PYCARD, PYD and CARD domain-containing; RFS, recurrence-free survival; RT-qPCR, quantitative PCR with reverse transcription; SF1, steroidogenic factor 1; THDOC, tetrahydrodeoxycorticosterone; THS, tetrahydro-11-deoxycortisol; ZWINT, ZW10-interacting kinetochore protein.

Review article

Glossary

Area under the receiver operating characteristic (ROC) curve

(AUC). A statistical analysis used to measure the diagnostic accuracy of a test. The y axis represents the sensitivity (true positive rate) and the x axis represents 1-specificity (true negative rate) of the test when using different cut-off values of the value measured. The closer the curve is to the left upper corner of the graph the more accurate the test is. Thus, a test with an AUC=1 is perfect and a test with an AUC=0.5 is completely random.

Classical subtype

The common histological subtype based on microscopic features of eosinophilic cytoplasm, often with thick fibrous bands and capsule, necrosis and mitotic figures.

Cushing syndrome

A disorder that is due to overproduction of cortisol over a prolonged period; also called hypercortisolism.

Decision curve analysis

A method of statistical analysis to evaluate prediction models, diagnostic tests and molecular markers.

Genomic imprinting

A mechanism of silencing of a gene in which the repressed allele is methylated and the active allele is unmethylated.

Helsinki scoring system

A system for diagnosis and prognosis based on evaluation of a combination of morphological (mitoses and necrosis) and immunohistochemical (Ki-67) parameters in patients.

Leydig cells

Cells in the testis that are the primary source of testosterone.

Lin-Weiss-Bisceglia system

A modified Weiss scoring system, which recommends that for oncocytic adrenocortical neoplasms, a malignancy can be indicated in the presence of one major criterion and indicated as uncertain in the presence of only minor criteria or considered as benign in the absence of both major and minor criteria.

Lynch syndrome

An inherited cancer syndrome that often genetically predisposes the patient to different types of cancer, especially colorectal cancer. Hence, it is also referred to as hereditary non-polyposis colorectal cancer.

Metabolic tumour volume

(MTV). A measurement of the metabolically active tumour volume based on tumour segmentation for the amount of 18F-fluorodeoxyglucose taken up on PET.

Myxoid subtype

When observed by microscopy, frequent cords or trabeculae of tumour cells appear floating in the stroma with diffuse pools or a lack of extracellular mucin. The tumour mucin is positive for Alcian blue.

Nomograms

Graphical calculating devices enabling an approximate graphical computation of a mathematical model predicting the relationship between variables and the probability of the outcome associated with those variables.

Oncocytic subtype

When observed by microscopy, abundant granular eosinophilic cytoplasm, excessive number of mitochondria and high-grade nuclear features are present. Frequent atypical mitotic figures and intranuclear inclusions are also observed.

Primary hyperaldosteronism

A disorder that is due to overproduction and release of aldosterone from the adrenal glands.

Reticulin algorithm

Distinguishes malignancy through an altered reticulin framework (a type of fibre in connective tissue composed of type III collagen in which these reticular fibres crosslink to form a fine meshwork) associated with either necrosis, a high mitotic rate or vascular invasion.

Sarcomatoid subtype

When observed by microscopy, frequent spindle tumour cells as well as giant cells are present. There is also prominent nuclear pleomorphism and atypical mitotic figures.

Scoring systems of Weiss et al.

The reference scoring method to distinguish between benign and malignant adrenocortical tumours in adults based on positive scores for features related to, for example, architecture, nucleus and the presence of any type of invasion, with each feature given a score of 1. A total score of 3 or more indicates a malignant tumour.

Standardized uptake value

(SUV). A semi-quantitative measure of the amount of 18F-fluorodeoxyglucose taken up by a tumour with PET. The value is determined by the ratio of activity per unit volume of a region of interest to the activity per unit whole body volume. SUVs are reported as the mean (the average over the region of interest) and the maximum (the highest in the region of interest).

Steroid sulfation

The sulfation of endogenous steroids. In general, sulfated steroids are not able to bind and activate their target

nuclear receptors and also require active transport into cells by an anion transporter as they are no longer lipophilic owing to the sulfation.

SV40 large T antigen

A dominant-acting oncoprotein derived from the polyomavirus SV40 and capable of inducing malignant transformation of various cell types. The transforming activity of this oncoprotein is largely due to its dysregulation of RB and p53.

Telomere maintenance pathway

The molecular pathway that regulates telemore length, which is essential for cancer cells to proliferate and not undergo senescence or apoptosis.

Total lesion glycolysis

(TLG). The product of the standardized uptake value and metabolic tumour volume, which more accurately reflects the glycolytic phenotype of a tumour and is associated with prognosis in several types of cancer.

Virilization

The acquisition of adult male physical features that develop in a female or young male, precociously, owing to excess androgen production.

Weighted correlation network analysis

(WGCNA). A widely used data mining method used especially for biological networks based on pairwise correlations between variables.

Wieneke criteria

A scoring system based on tumour size, local invasion and histological features, which distinguishes between benign and malignant tumours as well as predicts the prognosis of paediatric adrenocortical carcinomas.

that this approach could be relatively easily translated into the clinic to better predict prognosis and personalize adjuvant treatment decisions in high-risk patients if standard cut-off values can be established for the levels of steroid sulfation and SULT and STS enzyme immunostaining.

Taken together, these studies demonstrate the translational implica- tions of steroid and metabolomic profiling of adrenocortical carcinoma for refined diagnosis and prognostication, especially when integrated with known clinical diagnostic and prognostic factors.

Review article

There are a limited number of cell lines available for adrenocortical carcinoma research, and the currently available cell lines are not rep- resentative of all of the molecular subtypes of human adrenocortical carcinoma. Therefore, it has become necessary to establish new adreno- cortical carcinoma cell lines in which not only the molecular features but also the steroid and metabolic profiles are characterized to enable us to understand how these additional features mechanistically relate to disease and therapeutic outcomes. A recent study described the development and characterization of an adrenocortical carcinoma cell line TVBF-7 that carries a nonsense mutation in adenomatous polyposis coli (APC) and that exhibits autonomous cortisol secretion128. In this study, the TVBF-7 cell line was directly compared with the NCI-H295R and MUC-1 adrenocortical carcinoma cell lines with respect to the muta- tional status of established driver gene mutations, steroidogenic signal- ling and electrophysiological properties as well as secretion profiles. This analysis revealed patterns of genetic alterations and steroidogenic features similar to known pathophysiological adrenocortical carcinoma subtypes. In particular, the steroidogenic gene expression profiling (CYP11A1,CYP17A1,HSD3B2,HSD17B4,CYP21A2,CYP11B1,CYP11B2,MC2R and AT1R) and hormone secretion profiling (cortisol, aldosterone, dehy- droepiandrosterone (DHEA), DHEAS, testosterone, 17-OH progesterone and others) under basal and stimulated conditions (in the presence of growth factors that regulate steroidogenesis, including adrenocortico- tropic hormone (ACTH) and forskolin) revealed differential endocrine functionalities between the cell lines that reflect clinical heterogeneities observed in patients with adrenocortical carcinoma128. It is hoped that the establishment of additional adrenocortical carcinoma cell lines and their detailed characterization could be used for testing of new thera- peutics. Specifically, these cell lines might help to select the molecular subtypes of adrenocortical carcinoma that are more likely to respond to a particular therapy and assess candidate therapeutic agents that might be effective against the altered steroidogenesis pathways seen in these cell lines and reflected in patient tumours.

With respect to the steroid profiling of paediatric adrenocortical carcinomas, there have been a limited number of studies. However, one case report of a patient with paediatric adrenocortical carcinoma meas- ured the steroid metabolites and expression patterns of steroidogenic genes in their tumour. This revealed that testosterone and DHEAS were produced in abundance in this paediatric adrenocortical carcinoma but also that the combined steroidogenic features resembled that of fetal adrenal and Leydig cells129. Interestingly, patients with paediatric adrenocortical carcinoma are often diagnosed with virilization due to overproduction of androgen (mostly dihydrotestosterone (DHT))11. There are two pathways for DHT biosynthesis: the classic (the synthe- sis route from cholesterol to DHT) and the back-door (an alternative synthesis route for DHT that bypasses fetal testicular testosterone) pathways. To determine whether the back-door pathway contributes to virilization in paediatric adrenocortical carcinoma, Marti et al.130 studied seven children who had androgen-producing adrenocortical tumours using steroid profiling and immunohistochemical expression analyses. They found that all cases produced high levels of androgens through either the classic and/or back-door pathways with variable levels of steroid enzyme expression that were distinct between car- cinomas and adenomas130. This study suggests that higher androgen expression and secretion in paediatric adrenocortical tumours may result from dysregulated steroidogenesis and through multiple ster- oidogenesis pathways. Thus, future research into agents that target androgen overproduction should be considered for patients with adrenocortical carcinoma.

Adult adrenocortical carcinomas exhibit increased glucose con- sumption, as also seen in other cancer types, with upregulated glucose transporter (GLUT1) expression compared with benign tumours131. In addition, GLUT1 expression in tumour tissue is associated with worse survival in adrenocortical carcinoma131. Several investigators have shown how this metabolic feature of GLUT1 overexpression and increased glucose uptake in adrenocortical carcinoma can indicate disease aggressiveness by measuring standardized uptake value (SUV), metabolic tumour volume (MTV) and total lesion glycolysis (TLG) on 18F-fluorodeoxyglucose (FDG) positron emission tomography (PET) scans performed in the clinic, where high SUV, MTV and TLG levels are independently associated with worse overall survival in patients with adrenocortical carcinoma132,133. Such an in vivo, real-time assess- ment of the metabolic state of the tumour with a PET scan could help to individualize decisions on whether to start treatment earlier, and in assessing treatment responses earlier. Lastly, in a study of paedi- atric adrenocortical carcinoma, GLUT1 was found to be differentially expressed between benign and malignant tumours, with higher expression in malignant tumours, which was associated with shorter survival134. Consistent with adult adrenocortical carcinomas, this find- ing suggests metabolic remodelling during tumorigenesis towards a hyperglycolytic phenotype.

Translational multi-omic analysis

Genomic and epigenomic analyses of adrenocortical carcinoma have been performed using various multi-omic platforms that have shed light on the important regulatory pathways in this cancer type, as out- lined above, but also have allowed for the identification of a parsimoni- ous set of diagnostic and prognostic markers that could be translated into the clinic to optimize patient management. Assié et al.135 performed a meta-analysis of studies of the transcriptome, methylome, chromo- some alterations and mutational profiles of adrenocortical carcinoma. This revealed previously identified136 differential expression of BUB1B (encodes a component of the mitotic checkpoint) and PINK1 (encodes a mitochondrial serine and threonine protein kinase) and differential DNA methylation of PAX5, GSTP1, PYCARD and PAX6, between adenomas and carcinomas; these were all associated with prognosis. Hence, in this study the investigators focused on analysis of these genes as a targeted molecular classifier in a training and validation cohort. They found that the combined expression level of BUB1B and PINK1 was an accurate predictor of overall survival in patients with adrenocortical carcinoma. In another study, publicly available transcriptome data from TCGA adrenocortical carcinoma data set was used to classify adrenocortical carcinoma at the transcriptome level137. Analysis by uniform manifold approximation and projection (UMAP), a form of machine learning, identified two distinct groups: adrenocortical carcinoma-UMAP1 and adrenocortical carcinoma-UMAP2. Furthermore, application of another machine learning method, known as random-forest-based machine learning, revealed a set of potential novel gene markers with significant differential expression between the two clusters (includ- ing sterol O-acyltransferase 1 (SOAT1), which encodes an enzyme that catalyses the formation of fatty acid-cholesterol esters, and EIF2A1). The gene markers were then validated to be differentially expressed between benign and malignant tumours and associated with good prognosis in adrenocortical carcinoma. A final study used a collection of eight different methods, including weighted correlation network analysis (WGCNA), differentially expressed gene analysis, expression level comparison, protein-protein interaction network construction, survival analysis, ROC analysis and decision curve analysis, to identify

Review article

possible prognostic biomarkers for adrenocortical carcinoma using seven independent data sets138. The researchers identified nine hub genes (genes with high connectivity based on protein-protein interac- tion network construction and WGCNA) associated with prognosis in patients with adrenocortical carcinoma, and four cell cycle regulatory genes, which they classified as ‘meaningful prognostic biomarkers’ (ASPM, BIRC5, CCNB2 and CDK1) because they had high accuracy (≥80% accuracy in three separate analyses) in predicting survival. In addition, copy number variation and mutations in these meaningful prognostic biomarkers were closely related to overall survival rates in patients with adrenocortical carcinoma, and their expression was also associated with the degree of immune cell infiltration. Lastly, two nomograms (overall survival nomogram and disease-free survival nomogram) were constructed as a visual means to provide clinicians with an accurate and fast method for survival prediction. The nomogram for predicting overall survival used ASPM gene expression level and pathological stage, and the nomogram for predicting disease-free survival used ASPM and CDK1 gene expression levels and pathological stage.

Immune microenvironment

Adrenocortical carcinoma is considered a non-immunogenic or immune-deplete cancer. This is because low levels of immune infil- tration, lymphocyte fraction (based on DNA methylation analysis) and immune-related gene expression have been reported in adrenocorti- cal carcinoma when compared with other cancer types139-141. Tian and colleagues142 using the cell-type identification by estimating relative subsets of RNA transcripts (CIBERSORT) computational algorithm143 found that infiltrating immune cells in adrenocortical carcinoma mainly consist of T cells, natural killer cells, mast cells and macrophages and, in an analysis of T cells specifically (CD3+, CD4+, CD8+ and forkhead box protein P3 (FOXP3)+) in adrenocortical carcinoma samples, CD8+ lymphocytes were identified as being the most common T cell population144. The number of tumour-infiltrating lymphocytes has been shown to be associated with better overall survival and disease-free survival in adrenocortical carcinoma, and a lower number of T helper cells was found to be inversely correlated with adrenocortical carcino- mas with glucocorticoid excess (resulting from tumour pathologies, such as Cushing syndrome)142,144. In relation to the latter correlation, it is well recognized that glucocorticoids are immunosuppressive, acting on both circulating and tumour-infiltrating immune cells145.

Adrenocortical carcinomas have decreased or absent expression of major histocompatibility complex class II (MHC II) compared with normal adrenal cortex and adrenocortical adenoma146. The MHC II mol- ecules of the immune system process and present antigens to T cells. This antigen presentation either induces an immunological response or can be blunted through apoptosis that is induced by the FAS-FAS ligand (FASL) system147,148. Wolkersdörfer and colleagues149 studied the expression of MHC II and the FAS-FASL system in adrenocortical carcinoma and adenoma samples. They found MHC II antigen expres- sion in 27.7% of adrenocortical adenomas but none in adrenocortical carcinomas, and lower FAS receptor and increased FASL expression in adrenocortical carcinomas compared with adrenocortical adeno- mas. These findings suggest that immune evasion in adrenocorti- cal carcinoma may be due to a lack of MHC II antigen expression, in conjunction with downregulation of the FAS receptor and increased expression of FASL, which have been associated with cancer cell resist- ance to apoptosis and infiltrating T cell apoptosis, respectively150,151. In paediatric adrenocortical tumours, MHC II expression was higher in adrenocortical adenoma than in carcinoma, and higher expression

of MHC II in adrenocortical carcinoma was associated with better prognosis152.

In addition to studies evaluating tumour-infiltrating immune cells and MHC II expression, additional immunoregulatory molecules in adult adrenocortical carcinoma have been studied. Expression of the lipopolysaccharide (LPS) signalling molecules, Toll-like recep- tor 4 (TLR4), CD14 and MD2 (also known as lymphocyte antigen 96) in adrenocortical carcinoma tissue is lower than in normal adrenal tissue153. TLR4 has been shown to regulate immune evasion, cancer progression and resistance to chemotherapy154,155 while CD14 acts as a co-receptor for both TLR4 and MD2. Ectopic expression of TLR4 and CD14 in the adrenocortical carcinoma cell line NCI-H295R resulted in apoptosis153. Another immunomodulatory molecule CD276 (also known as B7-H3) is overexpressed in adrenocortical carcinoma and is associated with an increased recurrence rate and worse prognosis156. CD276 functions as an important immune checkpoint protein known to have a role in epithelial-to-mesenchymal transition (EMT) and progres- sion in other cancer types157,158 . Clinical trials using immune checkpoint inhibitors (PD1 and PDL1) in adrenocortical carcinoma have shown limited efficacy (6-23% partial response rate)159-164. Unlike other cancer types, there have been no consistent association between PD1 and PDL1 expression levels, tumour mutational burden and tumour-infiltrating lymphocyte number in adrenocortical tumour samples and response to immune checkpoint inhibitor treatment, but the study cohorts have been relatively small159-164.

Glucocorticoids impair immunity through several mechanisms145. Patients with adrenocortical carcinoma can have elevated levels of steroid precursors in their tumours, even when the serum cortisol lev- els are normal as a result of activation of the steroidogenesis pathway in tumours120,165 (Fig. 3). One of the main mechanisms of antitumour immunity is through T cell activation, and glucocorticoids can coun- teract this by directly inducing T cell apoptosis and decreasing secre- tion of the immunomodulatory cytokine IL-2 from immune cells145. As alluded to above, in adrenocortical carcinoma, excess glucocorticoids are associated with tumours that have lower T cell infiltration and worse prognosis144. Furthermore, in TCGA adrenocortical carcinoma study44, the patient group with a high steroid tumour profile (based on mRNA expression levels of 25 genes with high expression in the adult normal adrenal cortex and associated with adrenocortical differentiation, including steroidogenic enzymes, cholesterol transporters and their transcription factor, steroidogenic factor 1 (SF1)) had worse prognosis and the patient group with a non-high steroid tumour profile had higher expression of genes involved in immune pathways. Thus, strategies to decrease excess steroids in patients with adrenocortical carcinoma in combination with immunotherapy may be more effective and are being currently investigated in clinical trials166. In addition, an adrenocortical carcinoma antigen peptide vaccine in combination with a PD1 inhibitor is currently being evaluated in a clinical trial167.

Lastly, another element of the tumour microenvironment as a whole but not the immune microenvironment specifically, adipose stem cells, have been shown to increase adrenocortical carcinoma cell proliferation and invasiveness. In co-culture in vitro experi- ments using human adipose stem cells and an adrenocortical carci- noma cell line (NCI-H295R), adipose stem cells exhibited decreased maturation and intracellular lipid content in the presence of the adrenocortical carcinoma cells, as well as increased activation of the CXC- chemokine ligand 12 (CXCL12)-CXC-chemokine receptor 7 (CXCR7) signalling pathway, which was associated with increased cancer cell migration168.

Review article

Mouse models

The most commonly used mouse model of adrenocortical carcinoma has been the xenograft model in immunocompromised mice using immortalized mouse and human cell lines, human primary adrenocorti- cal carcinoma cells and patient-derived tumours. These models have been used to evaluate anticancer drugs169-172. Despite being commonly used, xenograft models have exhibited inconsistency in experimental outcomes, highlighting the need to standardize this model. For exam- ple, Hantel et al.173 found poor reproducibility using the adrenocortical carcinoma NCI-H295R cell line between two clones of the cell line that were authenticated by short-random repeat profiling. The xenograft engraftment rate, in vivo growth rate and tumour vascularity were different between the two cell line clones, as was their response to the current standard-of-care treatment etoposide, doxorubicin, cisplatin plus mitotane. Other disadvantages of xenograft models in general include a lack of being able to determine tumour immune responses of treatment regimens owing to the mice being immunocompromised in most cases. Furthermore, orthotopic models of adrenocortical car- cinoma have not been widely used as it is a technically difficult surgical procedure174. Therefore, there is an urgent need for the development of new mouse models with genetic alterations that have been identified in human adrenocortical carcinoma. The earliest engineered mouse models aiming to mimic human adrenocortical carcinoma were gener- ated by activation of the WNT-B-catenin signalling pathway and/or IGF2 overexpression. Interestingly, the mouse models with these genetic alterations resulted in the formation of adrenocortical carcinoma at very low rates after a prolonged period of time78,79,175. Recently, Batisse-Lignier and colleagues176 reported a transgenic mouse model with adrenocortical tissue-specific expression of SV40 large T antigen to determine whether inhibition of p53 and RB was oncogenic. They observed that the mice developed large adrenal tumours with activa- tion of the WNT-B-catenin signalling pathway and also exhibited liver and lung metastases. Furthermore, activation of the mTOR complex 1 (mTORC1) pathway was shown to be an early event of tumorigenesis in this model, and analysis of human adrenocortical carcinoma samples also showed activation of this pathway. Moreover, mTORC1 inhibi- tion, using rapamycin, inhibited tumour growth in this mouse model. Borges et al.177 have also developed a genetically engineered mouse model of adrenocortical carcinoma. In this mouse model, combined ß-catenin activation and Trp53 deletion results in the development of adrenocortical carcinoma with associated metastases. Interestingly, in another example of a mouse model, Znrf3 knockout in female mice results in the development of adrenocortical carcinoma with metas- tases late in life (78 weeks or 18 months and in 75% of cases) but not in male mice, where the appearance of adrenal hyperplasia regresses as a result of activation of senescence and an innate immune response to these senescent cells by phagocytic macrophages178. This suggests, along with other data179,180, that ZNRF3 functions as a tumour suppressor with differential activation of senescence and innate immune responses between the sexes, which may be due to androgen receptor activation and increased myeloid cell accumulation. Such mouse models will be an important resource for preclinical testing of candidate therapeutic compounds, especially of agents that target these specific altered signalling pathways (Table 1).

Several investigators have recently reported successful establish- ment of adult and paediatric patient-derived xenografts (PDXs) as well as a humanized PDX mouse model of adrenocortical carcinoma173,181-183. Kiseljak-Vassiliades and colleagues181 established two PDXs of adreno- cortical carcinoma with corresponding cell lines, which all retained

expression of adrenal cortex markers as well as exhibiting hormonal secretion profiles (such as secretion of cortisol) similar to those seen in samples of human tumours. The genetic alterations present in both of these models were known genetic alterations of human adrenocortical carcinoma: one PDX had a CTNNB1 mutation and the other had a TP53 mutation and loss of MSH2 consistent with the patient’s known pathol- ogy of Lynch syndrome. This same research group also established a humanized PDX mouse model by intravenous or intrahepatic injection of CD34+ cells that were harvested from human cord blood into mice, which were subsequently implanted with a patient-derived tumour182. They tested a PD1 immune checkpoint inhibitor in this model and found significant inhibition of tumour growth with increased numbers of tumour-infiltrating lymphocytes. Interestingly, the patient from which the PDX was derived was also treated with a PD1 immune checkpoint inhibitor and had an exceptional response similar to that observed in the humanized PDX mouse model. A PDX model generated from a paediatric patient with adrenocortical carcinoma resulting from a germline TP53G245C mutation had a molecular profile similar to that of the patient’s tumour; furthermore, the cytotoxic chemotherapies tested on the model gave a similar response profile with the tumour model being sensitive to topotecan treatment183. Based on their find- ings, the investigators used the schedule for topotecan treatment from the model in another paediatric patient who had developed recurrent metastatic adrenocortical carcinoma after initial surgical resection and chemotherapy with etoposide, doxorubicin, cisplatin plus mitotane, and topotecan treatment resulted in stable disease for the 4 months during which the patient received treatment. These studies demonstrate that PDX models can be used in a personalized approach to test current clinically available therapeutics that may be effective in individual patients as well as to test candidate agents that target specific altered pathways in adrenocortical carcinoma.

Conclusions

There has been improved understanding of the molecular pathways involved in adrenocortical carcinoma as a result of recent genomic, epigenetic and metabolomic studies. Some of the pathways involved in adrenocortical carcinoma such as IGF, DDR, WNT-ß-catenin, PI3K- AKT-mTOR and p53 are known to be pathogenic as they are also involved in other more common cancer types. However, whether the newly identified genomic alterations and changes in miRNA expres- sion and epigenetic changes that are associated with adrenocortical carcinoma are pathogenic and drive cancer progression is unclear and needs to be studied. To understand the role of these alterations, future mechanistic studies using preclinical models to evaluate whether they are drivers of adrenocortical carcinoma initiation and/or progression will be important, as targeting IGF signalling in clinical trials, using both small-molecule inhibitors and monoclonal antibodies69,70 has not yielded promising results and calls into question whether the IGF- IGF1R pathway is really a driver of disease progression and metastasis in adrenocortical carcinoma.

The integration of recent genomic, epigenetic and metabolomic data from the aforementioned studies into clinical and pathological evaluation has great potential to enhance diagnostic accuracy and prognostication, and to lead to the development of new therapeutics that target alterations present in the patient-specific tumour pro- file. Future translational studies that target specific altered pathways based on either studies of other common cancers or preclinical data specifically in adrenocortical carcinoma will be important to progress treatment options for patients with adrenocortical carcinoma.

Review article

Although multi-omic studies have resulted in definition of the common and unique molecular and biological features of adrenocor- tical carcinoma, the future development of preclinical models that fully recapitulate these unique molecular and biological features of human adrenocortical carcinoma will be key to identifying targeted drugs and assessing their efficacy, as well as for understanding the mechanisms of primary treatment resistance. Future research should also focus on refining our understanding of the tumour microenvi- ronment and more specifically the immune microenvironment of adrenocortical carcinomas to determine whether it could be altered to make it as immunogenic as other cancers. This will be important for identification of which patients could benefit from immunotherapy and which immunotherapeutic agents would be best as treatments for adrenocortical carcinoma. In this context, future research focusing on strategies to reduce the steroid-rich tumour microenvironment in adrenocortical carcinoma, which is known to be immunosuppressive, will be important.

Given that adrenocortical carcinoma is a rare cancer and drug development is a long and costly process, a future strategy for treating this cancer type might be the repurposing of drugs used for other indi- cations. Thus, the approach of quantitative high-throughout screening of known and clinically approved drugs for their efficacy in adreno- cortical carcinoma using in vitro and in vivo models that recapitulate the biology of adrenocortical carcinoma could circumvent the long process of studying the pharmacodynamic and pharmacokinetic prop- erties and toxicity of a novel drug before even beginning to conduct clinical trials. However, mechanistic studies of the drug targets for candidate repurposed drugs should also be performed to know which adrenocortical carcinoma molecular subtypes are likely to respond to the candidate agents. In addition, it is not likely that monotherapy in adrenocortical carcinoma will be effective, so screening for repurposed drugs should focus on matrix screening (two or more drugs in combina- tion) of compounds found to have anticancer activity as a single agent or in combination with drugs (for example, mitotane) that already have some efficacy in adrenocortical carcinoma. Such a strategy could help to identify drugs with synergistic or additive anticancer activity and thus reduce the effective dose concentration needed for anticancer activity and lower the drug side-effect profiles.

In summary, there have been advances in understanding the molecular pathways involved in adrenocortical carcinoma, and many studies have demonstrated that these molecular alterations alone or when integrated with clinical and pathological features can refine the diagnosis and prognostication of adrenocortical carcinoma. Future research focused on targeting molecular alterations in adrenocortical carcinoma, alteration of the immune tumour microenvironment and repurposing of existing drugs are needed to improve outcomes for patients with adrenocortical carcinoma.

Published online: 19 October 2023

References

1. Bilimoria, K. Y. et al. Adrenocortical carcinoma in the United States: treatment utilization and prognostic factors. Cancer 113, 3130-3136 (2008).

2. Else, T. et al. Adrenocortical carcinoma. Endocr. Rev. 35, 282-326 (2014).

3. Kebebew, E., Reiff, E., Duh, Q. Y., Clark, O. H. & McMillan, A. Extent of disease at presentation and outcome for adrenocortical carcinoma: have we made progress? World J. Surg. 30, 872-878 (2006).

4. Tella, S. H., Kommalapati, A., Yaturu, S. & Kebebew, E. Predictors of survival in adrenocortical carcinoma: an analysis from the national cancer database. J. Clin. Endocrinol. Metab. 103, 3566-3573 (2018).

5. Tierney, J. F. et al. National treatment practice for adrenocortical carcinoma: have they changed and have we made any progress? J. Clin. Endocrinol. Metab. 104, 5948-5956 (2019).

6. Kerkhofs, T. M. et al. Adrenocortical carcinoma: a population-based study on incidence and survival in the Netherlands since 1993. Eur. J. Cancer 49, 2579-2586 (2013).

7. McAteer, J. P., Huaco, J. A. & Gow, K. W. Predictors of survival in pediatric adrenocortical carcinoma: a Surveillance, Epidemiology, and End Results (SEER) program study. J. Pediatr. Surg. 48, 1025-1031 (2013).

8. Fassnacht, M. et al. Limited prognostic value of the 2004 International Union Against Cancer staging classification for adrenocortical carcinoma: proposal for a revised TNM classification. Cancer 115, 243-250 (2009).

9. Elhassan, Y. S. et al. S-GRAS score for prognostic classification of adrenocortical carcinoma: an international, multicenter ENSAT study. Eur. J. Endocrinol. 186, 25-36 (2021).

10. Beuschlein, F. et al. Major prognostic role of Ki67 in localized adrenocortical carcinoma after complete resection. J. Clin. Endocrinol. Metab. 100, 841-849 (2015).

11. Michalkiewicz, E. et al. Clinical and outcome characteristics of children with adrenocortical tumors: a report from the International Pediatric Adrenocortical Tumor Registry. J. Clin. Oncol. 22, 838-845 (2004).

12. Loncar, Z. et al. Survival and prognostic factors for adrenocortical carcinoma: a single institution experience. BMC Urol. 15, 43 (2015).

13. Li, P., Su, X., Zhang, X., Sun, L. & Zhang, G. Prognostic factors of adrenocortical carcinoma: experience from a Regional Medical Center in Eastern China. Int. J. Gen. Med. 16, 453-465 (2023).

14. Lughezzani, G. et al. The European Network for the Study of Adrenal Tumors staging system is prognostically superior to the International Union Against Cancer-staging system: a North American validation. Eur. J. Cancer 46, 713-719 (2010).

15. Terzolo, M. et al. Results of the ADIUVO study, the first randomized trial on adjuvant mitotane in adrenocortical carcinoma patients. J. Endocr. Soc. 5, A166-A167 (2021). Fassnacht, M. et al. Combination chemotherapy in advanced adrenocortical carcinoma. N. Engl. J. Med. 366, 2189-2197 (2012). 16.

17. Fassnacht, M. et al. European Society of Endocrinology Clinical Practice Guidelines on the management of adrenocortical carcinoma in adults, in collaboration with the European Network for the Study of Adrenal Tumors. Eur. J. Endocrinol. 179, G1-G46 (2018).

18. Challis, B. G. et al. Familial adrenocortical carcinoma in association with Lynch syndrome. J. Clin. Endocrinol. Metab. 101, 2269-2272 (2016).

19. Gatta-Cherifi, B. et al. Adrenal involvement in MEN1. Analysis of 715 cases from the Groupe d’Etude des Tumeurs Endocrines database. Eur. J. Endocrinol. 166, 269-279 (2012).

20. Hampel, H. et al. Screening for the Lynch syndrome (hereditary nonpolyposis colorectal cancer). N. Engl. J. Med. 352, 1851-1860 (2005).

21. Herrmann, L. J. et al. TP53 germline mutations in adult patients with adrenocortical carcinoma. J. Clin. Endocrinol. Metab. 97, E476-E485 (2012).

22. MacFarland, S. P. et al. Management of adrenal masses in patients with Beckwith-Wiedemann syndrome. Pediatr. Blood Cancer 64, 10.1002/pbc.26432 (2017).

23. . Pinto, E. M. et al. Identification of clinical and biologic correlates associated with outcome in children with adrenocortical tumors without germline TP53 mutations: a st jude adrenocortical tumor registry and children’s oncology group study. J. Clin. Oncol. 35, 3956-3963 (2017).

24. Raymond, V. M. et al. Prevalence of germline TP53 mutations in a prospective series of unselected patients with adrenocortical carcinoma. J. Clin. Endocrinol. Metab. 98, E119-E125 (2013).

25. 5. Raymond, V. M. et al. Adrenocortical carcinoma is a Lynch syndrome-associated cancer. J. Clin. Oncol. 31, 3012-3018 (2013).

26. Seki, M. et al. Loss of normal allele of the APC gene in an adrenocortical carcinoma from a patient with familial adenomatous polyposis. Hum. Genet. 89, 298-300 (1992).

27. Shiroky, J. S., Lerner-Ellis, J. P., Govindarajan, A., Urbach, D. R. & Devon, K. M. Characteristics of adrenal masses in familial adenomatous polyposis. Dis. Colon Rectum 61, 679-685 (2018).

28. 3. Wakatsuki, S. et al. Adrenocortical tumor in a patient with familial adenomatous polyposis: a case associated with a complete inactivating mutation of the APC gene and unusual histological features. Hum. Pathol. 29, 302-306 (1998).

29. Wasserman, J. D. et al. Prevalence and functional consequence of TP53 mutations in pediatric adrenocortical carcinoma: a Children’s Oncology Group study. J. Clin. Oncol. 33, 602-609 (2015).

30. Juhlin, C. C. et al. What did we learn from the molecular biology of adrenal cortical neoplasia? From histopathology to translational genomics. Endocr. Pathol. 32, 102-133 (2021).

31. Grisanti, S., Cosentini, D., Sigala, S. & Berruti, A. Molecular genotyping of adrenocortical carcinoma: a systematic analysis of published literature 2019-2021. Curr. Opin. Oncol. 34, 19-28 (2022).

32. Else, T. et al. Adrenocortical carcinoma and succinate dehydrogenase gene mutations: an observational case series. Eur. J. Endocrinol. 177, 439-444 (2017).

33. Grisanti, S. et al. 29MO Germline variants NGS characterization in patients with non-syndromic adrenocortical carcinoma. ESMO Open 8, 101050 (2023).

34. Custodio, G. et al. Impact of neonatal screening and surveillance for the TP53 R337H mutation on early detection of childhood adrenocortical tumors. J. Clin. Oncol. 31, 2619-2626 (2013).

35. Latronico, A. C. et al. An inherited mutation outside the highly conserved DNA-binding domain of the p53 tumor suppressor protein in children and adults with sporadic adrenocortical tumors. J. Clin. Endocrinol. Metab. 86, 4970-4973 (2001).

Review article

36. Pinto, E. M. et al. Founder effect for the highly prevalent R337H mutation of tumor suppressor p53 in Brazilian patients with adrenocortical tumors. Arq. Bras. Endocrinol. Metab. 48, 647-650 (2004).

37. Ribeiro, R. C. et al. An inherited p53 mutation that contributes in a tissue-specific manner to pediatric adrenal cortical carcinoma. Proc. Natl Acad. Sci. USA 98, 9330-9335 (2001).

38. Mete, O. et al. Overview of the 2022 who classification of adrenal cortical tumors. Endocr. Pathol. 33, 155-196 (2022).

39. . Minner, S., Schreiner, J. & Saeger, W. Adrenal cancer: relevance of different grading systems and subtypes. Clin. Transl. Oncol. 23, 1350-1357 (2021).

40. Weiss, L. M. Comparative histologic study of 43 metastasizing and nonmetastasizing adrenocortical tumors. Am. J. Surg. Pathol. 8, 163-169 (1984).

41. Weiss, L. M., Medeiros, L. J. & Vickery, A. L. Jr Pathologic features of prognostic significance in adrenocortical carcinoma. Am. J. Surg. Pathol. 13, 202-206 (1989).

42. Pittaway, J. F. H. & Guasti, L. Pathobiology and genetics of adrenocortical carcinoma. J. Mol. Endocrinol. 62, R105-R119 (2019).

43. Assie, G. et al. Integrated genomic characterization of adrenocortical carcinoma. Nat. Genet. 46, 607-612 (2014).

This study demonstrates that cataloguing the genomic changes in adrenocortical carcinoma reveals distinct molecular groups and that these are associated with prognosis.

14. Zheng, S. et al. Comprehensive pan-genomic characterization of adrenocortical carcinoma. Cancer Cell 29, 723-736 (2016).

This study comprehensively identifies genomic alterations associated with adult adrenocortical carcinoma and their association with patient outcome.

45. Sbiera, I. et al. Role of FGF receptors and their pathways in adrenocortical tumors and possible therapeutic implications. Front. Endocrinol. 12, 795116 (2021).

46. Tamburello, M. et al. FGF/FGFR signaling in adrenocortical development and tumorigenesis: novel potential therapeutic targets in adrenocortical carcinoma. Endocrine 77, 411-418 (2022).

47. Pereira, S. S. et al. IGF2 role in adrenocortical carcinoma biology. Endocrine 66, 326-337 (2019).

48. Pinto, E. M. et al. Genomic landscape of paediatric adrenocortical tumours. Nat. Commun. 6, 6302 (2015).

This study shows the genomic alterations that are prevalent in and unique to paediatric adrenocortical carcinomas.

49. Pozdeyev, N. et al. Targeted genomic analysis of 364 adrenocortical carcinomas. Endocr. Relat. Cancer 28, 671-681 (2021).

This study is one of the largest mutational analyses performed using NGS and identifies novel genomic alterations that could be targeted for therapy.

50. Hoadley, K. A. et al. Cell-of-origin patterns dominate the molecular classification of 10,000 tumors from 33 types of cancer. Cell 173, 291-304.e6 (2018).

51. Knijnenburg, T. A. et al. Genomic and molecular landscape of DNA damage repair deficiency across the cancer genome atlas. Cell Rep. 23, 239-254.e6 (2018).

52. Sondka, Z. et al. The COSMIC cancer gene census: describing genetic dysfunction across all human cancers. Nat. Rev. Cancer 18, 696-705 (2018).

53. Kurtz, A. et al. Somatic mitochondrial DNA mutations in neurofibromatosis type 1-associated tumors. Mol. Cancer Res. 2, 433-441 (2004).

54. Wang, Q. et al. Neurofibromatosis type 1 gene as a mutational target in a mismatch repair-deficient cell type. Hum. Genet. 112, 117-123 (2003).

55. Santos, M. A. et al. DNA-damage-induced differentiation of leukaemic cells as an anti-cancer barrier. Nature 514, 107-111 (2014).

56. Pemov, A., Park, C., Reilly, K. M. & Stewart, D. R. Evidence of perturbations of cell cycle and DNA repair pathways as a consequence of human and murine NF1-haploinsufficiency. BMC Genomics 11, 194 (2010).

57. Bielski, C. M. et al. Genome doubling shapes the evolution and prognosis of advanced cancers. Nat. Genet. 50, 1189-1195 (2018).

58. Letouze, E. et al. SNP array profiling of childhood adrenocortical tumors reveals distinct pathways of tumorigenesis and highlights candidate driver genes. J. Clin. Endocrinol. Metab. 97, E1284-E1293 (2012).

59. Zack, T. I. et al. Pan-cancer patterns of somatic copy number alteration. Nat. Genet. 45, 1134-1140 (2013).

60. Juhlin, C. C. et al. Whole-exome sequencing characterizes the landscape of somatic mutations and copy number alterations in adrenocortical carcinoma. J. Clin. Endocrinol. Metab. 100, E493-E502 (2015).

61. Gicquel, C. et al. Rearrangements at the 11p15 locus and overexpression of insulin-like growth factor-Il gene in sporadic adrenocortical tumors. J. Clin. Endocrinol. Metab. 78, 1444-1453 (1994).

62. Gicquel, C. et al. Structural and functional abnormalities at 11p15 are associated with the malignant phenotype in sporadic adrenocortical tumors: study on a series of 82 tumors. J. Clin. Endocrinol. Metab. 82, 2559-2565 (1997).

63. Peixoto Lira, R. C. et al. IGF2 and IGF1R in pediatric adrenocortical tumors: roles in metastasis and steroidogenesis. Endocr. Relat. Cancer 23, 481-493 (2016).

64. Guillaud-Bataille, M. et al. IGF2 promotes growth of adrenocortical carcinoma cells, but its overexpression does not modify phenotypic and molecular features of adrenocortical carcinoma. PLOS ONE 9, e103744 (2014).

65. Nielsen, H. M. et al. Copy number variations alter methylation and parallel IGF2 overexpression in adrenal tumors. Endocr. Relat. Cancer 22, 953-967 (2015).

66. Rosati, R. et al. High frequency of loss of heterozygosity at 11p15 and IGF2 overexpression are not related to clinical outcome in childhood adrenocortical tumors positive for the R337H TP53 mutation. Cancer Genet. Cytogenet. 186, 19-24 (2008).

67. Szabo, P. E., Tang, S. H., Silva, F. J., Tsark, W. M. & Mann, J. R. Role of CTCF binding sites in the Igf2/H19 imprinting control region. Mol. Cell Biol. 24, 4791-4800 (2004).

68. Barlaskar, F. M. et al. Preclinical targeting of the type I insulin-like growth factor receptor in adrenocortical carcinoma. J. Clin. Endocrinol. Metab. 94, 204-212 (2009).

69. US National Library of Medicine. ClinicalTrials.gov https://www.clinicaltrials.gov/study/ NCT00831844 (2015).

70. Fassnacht, M. et al. Linsitinib (OSI-906) versus placebo for patients with locally advanced or metastatic adrenocortical carcinoma: a double-blind, randomised, phase 3 study. Lancet Oncol. 16, 426-35 (2015).

71. Sasano, H., Satoh, F. & Nakamura, Y. Roles of the pathologist in evaluating surrogate markers for medical therapy in adrenocortical carcinoma. Endocr. Pathol. 25, 366-370 (2014).

72. Adam, P. et al. Epidermal growth factor receptor in adrenocortical tumors: analysis of gene sequence, protein expression and correlation with clinical outcome. Mod. Pathol. 23, 1596-1604 (2010).

73. Quinkler, M. et al. Treatment of advanced adrenocortical carcinoma with erlotinib plus gemcitabine. J. Clin. Endocrinol. Metab. 93, 2057-2062 (2008).

74. Ardolino, L., Hansen, A., Ackland, S. & Joshua, A. Advanced adrenocortical carcinoma (ACC): a review with focus on second-line therapies. Horm. Cancer 11, 155-169 (2020).

75. Boumahdi, S. & de Sauvage, F. J. The great escape: tumour cell plasticity in resistance to targeted therapy. Nat. Rev. Drug Discov. 19, 39-56 (2020).

76. Quintanal-Villalonga, A. et al. Lineage plasticity in cancer: a shared pathway of therapeutic resistance. Nat. Rev. Clin. Oncol. 17, 360-371 (2020).

77. Gara, S. K. et al. Metastatic adrenocortical carcinoma displays higher mutation rate and tumor heterogeneity than primary tumors. Nat. Commun. 9, 4172 (2018).

This study demonstrates that genomic alterations between primary adrenocortical carcinoma and recurrent and metastatic adrenocortical carcinoma are different with the latter exhibiting more tumour heterogeneity.

78. Heaton, J. H. et al. Progression to adrenocortical tumorigenesis in mice and humans through insulin-like growth factor 2 and ß-catenin. Am. J. Pathol. 181, 1017-1033 (2012). 79. Berthon, A. et al. Constitutive beta-catenin activation induces adrenal hyperplasia and promotes adrenal cancer development. Hum. Mol. Genet. 19, 1561-1576 (2010).

80. Lin, S. & Gregory, R. I. MicroRNA biogenesis pathways in cancer. Nat. Rev. Cancer 15, 321-333 (2015).

81. Huarte, M. The emerging role of IncRNAs in cancer. Nat. Med. 21, 1253-1261 (2015).

82. Liu, J., Kahri, A. I., Heikkila, P., Ilvesmaki, V. & Voutilainen, R. H19 and insulin-like growth factor-Il gene expression in adrenal tumors and cultured adrenal cells. J. Clin. Endocrinol. Metab. 80, 492-496 (1995).

83. Hatada, I. & Mukai, T. Genomic imprinting of p57KIP2, a cyclin-dependent kinase inhibitor, in mouse. Nat. Genet. 11, 204-206 (1995).

84. Liu, J., Kahri, A. I., Heikkila, P. & Voutilainen, R. Ribonucleic acid expression of the clustered imprinted genes, p57KIP2, insulin-like growth factor II, and H19, in adrenal tumors and cultured adrenal cells. J. Clin. Endocrinol. Metab. 82, 1766-1771 (1997).

85. Soon, P. S. et al. miR-195 and miR-483-5p identified as predictors of poor prognosis in adrenocortical cancer. Clin. Cancer Res. 15, 7684-7692 (2009).

86. Feinmesser, M. et al. Specific microRNAs differentiate adrenocortical adenomas from carcinomas and correlate with Weiss histopathologic system. Appl. Immunohistochem. Mol. Morphol. 23, 522-531 (2015).

87. Koperski, Ł. et al. Next-generation sequencing reveals microRNA markers of adrenocortical tumors malignancy. Oncotarget 8, 49191-49200 (2017).

88. Chehade, M., Bullock, M., Glover, A., Hutvagner, G. & Sidhu, S. Key microRNA’s and their targetome in adrenocortical cancer. Cancers 12, 2198 (2020).

89. Mitchell, P. S. et al. Circulating microRNAs as stable blood-based markers for cancer detection. Proc. Natl Acad. Sci. USA 105, 10513-10518 (2008).

90. Chabre, O. et al. Serum miR-483-5p and miR-195 are predictive of recurrence risk in adrenocortical cancer patients. Endocr. Relat. Cancer 20, 579-594 (2013).

91. Szabó, D. R. et al. Analysis of circulating microRNAs in adrenocortical tumors. Lab. Invest. 94, 331-339 (2014).

92. Patel, D. et al. MiR-34a and miR-483-5p are candidate serum biomarkers for adrenocortical tumors. Surgery 154, 1224-1228 (2013).

93. 3. Wu, Y. et al. MicroRNA-205 suppresses the growth of adrenocortical carcinoma SW-13 cells via targeting Bcl-2. Oncol. Rep. 34, 3104-3110 (2015).

94. Jain, M. et al. ZNF367 inhibits cancer progression and is targeted by miR-195. PLOS ONE 9, e101423 (2014).

95. Kwok, G. T. Y. et al. microRNA-431 as a chemosensitizer and potentiator of drug activity in adrenocortical carcinoma. Oncologist 24, e241-e250 (2019).

6. Hassan, N., Zhao, J. T., Glover, A., Robinson, B. G. & Sidhu, S. B. Reciprocal interplay of miR-497 and MALAT1 promotes tumourigenesis of adrenocortical cancer. Endocr. Relat. Cancer 26, 677-688 (2019).

97. Kalinowski, F. C. et al. microRNA-7: a tumor suppressor miRNA with therapeutic potential. Int. J. Biochem. Cell Biol. 54, 312-317 (2014).

98. Gara, S. K. et al. Integrated genome-wide analysis of genomic changes and gene regulation in human adrenocortical tissue samples. Nucleic Acids Res. 43, 9327-9339 (2015).

99. Glover, A. R. et al. Long noncoding RNA profiles of adrenocortical cancer can be used to predict recurrence. Endocr. Relat. Cancer 22, 99-109 (2015).

One of the first studies to show that the IncRNA expression profile of adrenocortical carcinoma is different from that of benign adrenocortical carcinoma and normal adrenal cortex.

Review article

100. Buishand, F. O. et al. Adrenocortical tumors have a distinct, long, non-coding RNA expression profile and LINC00271 is downregulated in malignancy. Surgery 167, 224-232 (2020).

101. Long, B. et al. Long noncoding RNA ASB16-AS1 inhibits adrenocortical carcinoma cell growth by promoting ubiquitination of RNA-binding protein HuR. Cell Death Dis. 11, 995 (2020).

102. Guo, N., Sun, Q., Fu, D. & Zhang, Y. Long non-coding RNA UCA1 promoted the growth of adrenocortical cancer cells via modulating the miR-298-CDK6 axis. Gene 703, 26-34 (2019).

103. Tombol, Z. et al. Integrative molecular bioinformatics study of human adrenocortical tumors: microRNA, tissue-specific target prediction, and pathway analysis. Endocr. Relat. Cancer 16, 895-906 (2009).

104. Duregon, E. et al. MicroRNA expression patterns in adrenocortical carcinoma variants and clinical pathologic correlations. Hum. Pathol. 45, 1555-1562 (2014).

105. Ozata, D. M. et al. The role of microRNA deregulation in the pathogenesis of adrenocortical carcinoma. Endocr. Relat. Cancer 18, 643-655 (2011).

106. Patterson, E. E., Holloway, A. K., Weng, J., Fojo, T. & Kebebew, E. MicroRNA profiling of adrenocortical tumors reveals miR-483 as a marker of malignancy. Cancer 117, 1630-1639 (2011).

107. Ettaieb, M., Kerkhofs, T., van Engeland, M. & Haak, H. Past, present and future of epigenetics in adrenocortical carcinoma. Cancers 12, 1218 (2020).

108. Rechache, N. S. et al. DNA methylation profiling identifies global methylation differences and markers of adrenocortical tumors. J. Clin. Endocrinol. Metab. 97, E1004-E1013 (2012).

109. Creemers, S. G. et al. Methylation of IGF2 regulatory regions to diagnose adrenocortical carcinomas. Endocr. Relat. Cancer 23, 727-737 (2016).

110. Barreau, O. et al. Identification of a CpG island methylator phenotype in adrenocortical carcinomas. J. Clin. Endocrinol. Metab. 98, E174-E184 (2013).

111. Jouinot, A. et al. DNA methylation is an independent prognostic marker of survival in adrenocortical cancer. J. Clin. Endocrinol. Metab. 102, 923-932 (2017).

112. Lippert, J. et al. Prognostic role of targeted methylation analysis in paraffin-embedded samples of adrenocortical carcinoma. J. Clin. Endocrinol. Metab. 107, 2892-2899 (2022).

113. Mohan, D. R. et al. Targeted assessment of GOS2 methylation identifies a rapidly recurrent, routinely fatal molecular subtype of adrenocortical carcinoma. Clin. Cancer Res. 25, 3276-3288 (2019).

114. Clay, M. R. et al. DNA methylation profiling reveals prognostically significant groups in pediatric adrenocortical tumors: a report from the International Pediatric Adrenocortical Tumor Registry. JCO Precis. Oncol. 3, PO.19.00163 (2019).

115. Drelon, C. et al. EZH2 is overexpressed in adrenocortical carcinoma and is associated with disease progression. Hum. Mol. Genet. 25, 2789-2800 (2016).

116. Tabbal, H. et al. EZH2 cooperates with E2F1 to stimulate expression of genes involved in adrenocortical carcinoma aggressiveness. Br. J. Cancer 121, 384-394 (2019).

117. Kebebew, E. Adrenal incidentaloma. N. Engl. J. Med. 384, 1542-1551 (2021).

118. Rege, J., Turcu, A. F., Else, T., Auchus, R. J. & Rainey, W. E. Steroid biomarkers in human adrenal disease. J. Steroid Biochem. Mol. Biol. 190, 273-280 (2019).

119. Taylor, D. R. et al. A 13-steroid serum panel based on LC-MS/MS: use in detection of adrenocortical carcinoma. Clin. Chem. 63, 1836-1846 (2017).

Along with Rege et al. (2019), this study shows the utility of measuring levels of steroids and their metabolities as biomarkers of adrenocortical carcinoma.

120. Chortis, V. et al. Urine steroid metabolomics as a novel tool for detection of recurrent adrenocortical carcinoma. J. Clin. Endocrinol. Metab. 105, e307-e318 (2020).

121. Suzuki, S. et al. Steroid metabolites for diagnosing and predicting clinicopathological features in cortisol-producing adrenocortical carcinoma. BMC Endocr. Disord. 20, 173 (2020).

122. Bancos, I. et al. Urine steroid metabolomics for the differential diagnosis of adrenal incidentalomas in the EURINE-ACT study: a prospective test validation study. Lancet Diabetes Endocrinol. 8, 773-781 (2020).

123. Patel, D. et al. Unique and novel urinary metabolomic features in malignant versus benign adrenal neoplasms. Clin. Cancer Res. 23, 5302-5310 (2017).

This study shows that patients with adrenocortical carcinoma have a unique urinary metabolome, which derives from the tumour tissue.

124. Mathe, E. A. et al. Noninvasive urinary metabolomic profiling identifies diagnostic and prognostic markers in lung cancer. Cancer Res. 74, 3259-3270 (2014).

125. Haznadar, M. et al. Urinary metabolites diagnostic and prognostic of intrahepatic cholangiocarcinoma. Cancer Epidemiol. Biomark. Prev. 28, 1704-1711 (2019).

126. Foster, P. A. & Mueller, J. W. SULFATION PATHWAYS: insights into steroid sulfation and desulfation pathways. J. Mol. Endocrinol. 61, T271-T283 (2018).

127. Sun, N. et al. Prognostic relevance of steroid sulfation in adrenocortical carcinoma revealed by molecular phenotyping using high-resolution mass spectrometry imaging. Clin. Chem. 65, 1276-1286 (2019).

128. Sigala, S. et al. A comprehensive investigation of steroidogenic signaling in classical and new experimental cell models of adrenocortical carcinoma. Cells 11, 1439 (2022).

129. Fujisawa, Y. et al. Combined steroidogenic characters of fetal adrenal and Leydig cells in childhood adrenocortical carcinoma. J. Steroid Biochem. Mol. Biol. 159, 86-93 (2016).

130. Marti, N. et al. Androgen production in pediatric adrenocortical tumors may occur via both the classic and/or the alternative backdoor pathway. Mol. Cell Endocrinol. 452, 64-73 (2017).

131. Fenske, W. et al. Glucose transporter GLUT1 expression is an stage-independent predictor of clinical outcome in adrenocortical carcinoma. Endocr. Relat. Cancer 16, 919-928 (2009).

132. Satoh, K., Patel, D., Dieckmann, W., Nilubol, N. & Kebebew, E. Whole body metabolic tumor volume and total lesion glycolysis predict survival in patients with adrenocortical carcinoma. Ann. Surg. Oncol. 22 (Suppl. 3), S714-S720 (2015).

133. Wrenn, S. M. et al. Higher SUV max on FDG-PET is associated with shorter survival in adrenocortical carcinoma. Am. J. Surg. 225, 309-314 (2023).

134. Pinheiro, C. et al. GLUT1 expression in pediatric adrenocortical tumors: a promising candidate to predict clinical behavior. Oncotarget 8, 63835-63845 (2017).

135. Assié, G. et al. Value of molecular classification for prognostic assessment of adrenocortical carcinoma. JAMA Oncol. 5, 1440-1447 (2019).

136. de Reyniès, A. et al. Gene expression profiling reveals a new classification of adrenocortical tumors and identifies molecular predictors of malignancy and survival. J. Clin. Oncol. 27, 1108-1115 (2009).

137. Marquardt, A. et al. Identifying new potential biomarkers in adrenocortical tumors based on mrna expression data using machine learning. Cancers 13, 4671 (2021).

138. Yi, X. et al. Identification of four novel prognostic biomarkers and construction of two nomograms in adrenocortical carcinoma: a multi-omics data study via bioinformatics and machine learning methods. Front. Mol. Biosci. 9, 878073 (2022).

139. Paré, L. et al. Association between PD1 mRNA and response to anti-PD1 monotherapy across multiple cancer types. Ann. Oncol. 29, 2121-2128 (2018).

140. Thorsson, V. et al. The immune landscape of cancer. Immunity 48, 812-830.e14 (2018).

141. Mohan, D. R., Lerario, A. M. & Hammer, G. D. Therapeutic targets for adrenocortical carcinoma in the genomics era. J. Endocr. Soc. 2, 1259-1274 (2018).

142. Tian, X. et al. Identification of tumor-infiltrating immune cells and prognostic validation of tumor-infiltrating mast cells in adrenocortical carcinoma: results from bioinformatics and real-world data. Oncoimmunology 9, 1784529 (2020).

143. Newman, A. M. et al. Robust enumeration of cell subsets from tissue expression profiles. Nat. Methods 12, 453-457 (2015).

144. Landwehr, L. S. et al. Interplay between glucocorticoids and tumor-infiltrating lymphocytes on the prognosis of adrenocortical carcinoma. J. Immunother. Cancer 8, e000469 (2020).

145. Coutinho, A. E. & Chapman, K. E. The anti-inflammatory and immunosuppressive effects of glucocorticoids, recent developments and mechanistic insights. Mol. Cell Endocrinol. 335, 2-13 (2011).

146. Marx, C., Wolkersdorfer, G. W., Brown, J. W., Scherbaum, W. A. & Bornstein, S. R. MHC class Il expression-a new tool to assess dignity in adrenocortical tumours. J. Clin. Endocrinol. Metab. 81, 4488-4491 (1996).

147. Ozdemirli, M. et al. Fas (CD95)/Fas ligand interactions regulate antigen-specific, major histocompatibility complex-restricted T/B cell proliferative responses. Eur. J. Immunol. 26, 415-419 (1996).

148. Brunner, T. et al. Cell-autonomous Fas (CD95)/Fas-ligand interaction mediates activation-induced apoptosis in T-cell hybridomas. Nature 373, 441-444 (1995).

149. Wolkersdörfer, G. W. et al. Prevalence of HLA-DRB1 genotype and altered Fas/Fas ligand expression in adrenocortical carcinoma. J. Clin. Endocrinol. Metab. 90, 1768-1774 (2005).

150. Hahne, M. et al. Melanoma cell expression of Fas(Apo-1/CD95) ligand: implications for tumor immune escape. Science 274, 1363-1366 (1996).

151. Xiao, W. et al. Loss of fas expression and function is coupled with colon cancer resistance to immune checkpoint inhibitor immunotherapy. Mol. Cancer Res. 17, 420-430 (2019).

152. Pinto, E. M. et al. Prognostic significance of major histocompatibility complex class ii expression in pediatric adrenocortical tumors: a St. Jude and Children’s Oncology Group Study. Clin. Cancer Res. 22, 6247-6255 (2016).

153. Kanczkowski, W. et al. Abrogation of TLR4 and CD14 expression and signaling in human adrenocortical tumors. J. Clin. Endocrinol. Metab. 95, E421-E429 (2010).

154. Huang, B. et al. Toll-like receptors on tumor cells facilitate evasion of immune surveillance. Cancer Res. 65, 5009-5014 (2005).

155. Szajnik, M. et al. TLR4 signaling induced by lipopolysaccharide or paclitaxel regulates tumor survival and chemoresistance in ovarian cancer. Oncogene 28, 4353-4363 (2009).

156. Liang, J. et al. Clinicopathological and prognostic characteristics of CD276 (B7-H3) expression in adrenocortical carcinoma. Dis. Markers 2020, 5354825 (2020).

157. Hofmeyer, K. A., Ray, A. & Zang, X. The contrasting role of B7-H3. Proc. Natl Acad. Sci. USA 105, 10277-10278 (2008).

158. Liao, H., Ding, M., Zhou, N., Yang, Y. & Chen, L. B7-H3 promotes the epithelial-mesenchymal transition of NSCLC by targeting SIRT1 through the PI3K/AKT pathway. Mol. Med. Rep. 25, 79 (2022).

159. Raj, N. et al. PD-1 blockade in advanced adrenocortical carcinoma. J. Clin. Oncol. 38, 71-80 (2020).

160. Habra, M. A. et al. Phase II clinical trial of pembrolizumab efficacy and safety in advanced adrenocortical carcinoma. J. Immunother. Cancer 7, 253 (2019).

161. Head, L. et al. Response to immunotherapy in combination with mitotane in patients with metastatic adrenocortical cancer. J. Endocr. Soc. 3, 2295-2304 (2019).

162. Bedrose, S. et al. Combined lenvatinib and pembrolizumab as salvage therapy in advanced adrenal cortical carcinoma. J. Immunother. Cancer 8, e001009 (2020).

163. Carneiro, B. A. et al. Nivolumab in metastatic adrenocortical carcinoma: results of a phase 2 trial. J. Clin. Endocrinol. Metab. 104, 6193-6200 (2019).

164. McGregor, B. A. et al. Results of a multicenter, phase 2 study of nivolumab and ipilimumab for patients with advanced rare genitourinary malignancies. Cancer 127, 840-849 (2021).

165. Grondal, S., Eriksson, B., Hagenas, L., Werner, S. & Curstedt, T. Steroid profile in urine: a useful tool in the diagnosis and follow up of adrenocortical carcinoma. Acta Endocrinol. 122, 656-663 (1990).

Review article

166. US National Library of Medicine. ClinicalTrials.gov https://www.clinicaltrials.gov/study/ NCT04373265 (2023).

167. US National Library of Medicine. ClinicalTrials.gov https://www.clinicaltrials.gov/study/ NCT04187404 (2023).

168. Armignacco, R. et al. The adipose stem cell as a novel metabolic actor in adrenocortical carcinoma progression: evidence from an in vitro tumor microenvironment crosstalk model. Cancers 11, 1931 (2019).

169. Jain, M. et al. TOP2A is overexpressed and is a therapeutic target for adrenocortical carcinoma. Endocr. Relat. Cancer 20, 361-370 (2013).

170. Satoh, K. et al. Identification of niclosamide as a novel anticancer agent for adrenocortical carcinoma. Clin. Cancer Res. 22, 3458-3466 (2016).

171. Jain, M. et al. Interleukin-13 receptor alpha2 is a novel therapeutic target for human adrenocortical carcinoma. Cancer 118, 5698-5708 (2012).

172. Barton, D. E., Foellmer, B. E., Wood, W. I. & Francke, U. Chromosome mapping of the growth hormone receptor gene in man and mouse. Cytogenet. Cell Genet. 50, 137-141 (1989).

173. Hantel, C. et al. Targeting heterogeneity of adrenocortical carcinoma: evaluation and extension of preclinical tumor models to improve clinical translation. Oncotarget 7, 79292-79304 (2016).

174. Cardoso, C. C., Bornstein, S. R. & Hornsby, P. J. Optimizing orthotopic cell transplantation in the mouse adrenal gland. Cell Transpl. 19, 565-572 (2010).

175. Ruggiero, C., Doghman-Bouguerra, M. & Lalli, E. How good are the current models of adrenocortical carcinoma for novel drug discovery? Expert Opin. Drug Discov. 17, 211-213 (2022).

176. Batisse-Lignier, M. et al. p53/Rb inhibition induces metastatic adrenocortical carcinomas in a preclinical transgenic model. Oncogene 36, 4445-4456 (2017).

This study demonstrates that known genetic alterations in human adrenocortical carcinoma introduced into a mouse can cause adrenocortical carcinoma similar in phenotype and histology to those human tumours and that this mouse model can in turn be used to evaluate the efficacy of targeted therapies.

177. Borges, K. S. et al. Wnt/ß-catenin activation cooperates with loss of p53 to cause adrenocortical carcinoma in mice. Oncogene 39, 5282-5291 (2020).

178. Warde, K. M. et al. Senescence-induced immune remodeling facilitates metastatic adrenal cancer in a sex-dimorphic manner. Nat. Aging 3, 846-865 (2023).

179. Wilmouth, J. J. Jr. et al. Sexually dimorphic activation of innate antitumor immunity prevents adrenocortical carcinoma development. Sci. Adv. 8, eadd0422 (2022).

180. Lyraki, R. et al. Crosstalk between androgen receptor and WNT/ß-catenin signaling causes sex-specific adrenocortical hyperplasia in mice. Dis. Model. Mech. 16, dmm050053 (2023).

181. Kiseljak-Vassiliades, K. et al. Development of new preclinical models to advance adrenocortical carcinoma research. Endocr. Relat. Cancer 25, 437-451 (2018).

182. Lang, J. et al. Development of an adrenocortical cancer humanized mouse model to characterize anti-pd1 effects on tumor microenvironment. J. Clin. Endocrinol. Metab. 105, 26-42 (2020).

This study develops a humanized adrenocortical carcinoma PDX mouse model to assess immunotherapy responses.

183. Pinto, E. M. et al. Establishment and characterization of the first pediatric adrenocortical carcinoma xenograft model identifies topotecan as a potential chemotherapeutic agent. Clin. Cancer Res. 19, 1740-1747 (2013).

184. Fiorentini, C. et al. Palbociclib inhibits proliferation of human adrenocortical tumor cells. Endocrine 59, 213-217 (2018).

185. Liang, R. et al. Targeted gene expression profile reveals CDK4 as therapeutic target for selected patients with adrenocortical carcinoma. Front. Endocrinol. 11, 219 (2020).

186. Hadjadj, D. et al. A hypothesis-driven approach identifies CDK4 and CDK6 inhibitors as candidate drugs for treatments of adrenocortical carcinomas. Aging 9, 2695-2716 (2017).

187. Nilubol, N. et al. Synergistic combination of flavopiridol and carfilzomib targets commonly dysregulated pathways in adrenocortical carcinoma and has biomarkers of response. Oncotarget 9, 33030-33042 (2018).

188. Doghman, M. et al. Regulation of insulin-like growth factor-mammalian target of rapamycin signaling by microRNA in childhood adrenocortical tumors. Cancer Res. 70, 4666-4675 (2010).

189. Doghman, M., Cazareth, J. & Lalli, E. The T cell factor/beta-catenin antagonist PKF115-584 inhibits proliferation of adrenocortical carcinoma cells. J. Clin. Endocrinol. Metab. 93, 3222-3225 (2008).

190. Gaujoux, S. et al. Silencing mutated ß-catenin inhibits cell proliferation and stimulates apoptosis in the adrenocortical cancer cell line H295R. PLOS ONE 8, e55743 (2013).

191. Cerquetti, L. et al. Rosiglitazone induces autophagy in H295R and cell cycle deregulation in SW13 adrenocortical cancer cells. Exp. Cell Res. 317, 1397-1410 (2011).

192. Cantini, G. et al. Rosiglitazone inhibits adrenocortical cancer cell proliferation by interfering with the IGF-IR intracellular signaling. PPAR Res. 2008, 904041 (2008).

193. Ferruzzi, P. et al. Thiazolidinediones inhibit growth and invasiveness of the human adrenocortical cancer cell line H295R. J. Clin. Endocrinol. Metab. 90, 1332-1339 (2005).

194. Sirianni, R. et al. Targeting estrogen receptor-a reduces adrenocortical cancer (ACC) cell growth in vitro and in vivo: potential therapeutic role of selective estrogen receptor modulators (SERMs) for ACC treatment. J. Clin. Endocrinol. Metab. 97, E2238-E2250 (2012).

195. Chimento, A. et al. GPER agonist G-1 decreases adrenocortical carcinoma (ACC) cell growth in vitro and in vivo. Oncotarget 6, 19190-19203 (2015).

196. Casaburi, I. et al. Estrogen related receptor a (ERRa) a promising target for the therapy of adrenocortical carcinoma (ACC). Oncotarget 6, 25135-25148 (2015).

197. Tamburello, M. et al. Preclinical evidence of progesterone as a new pharmacological strategy in human adrenocortical carcinoma cell lines. Int. J. Mol. Sci. 24, 6829 (2023).

198. Doghman, M. et al. Inhibition of adrenocortical carcinoma cell proliferation by steroidogenic factor-1 inverse agonists. J. Clin. Endocrinol. Metab. 94, 2178-2183 (2009).

199. Cheng, Y., Kerppola, R. E. & Kerppola, T. K. ATR-101 disrupts mitochondrial functions in adrenocortical carcinoma cells and in vivo. Endocr. Relat. Cancer 23, 1-19 (2016).

200. LaPensee, C. R. et al. ATR-101, a selective and potent inhibitor of acyl-CoA acyltransferase 1, induces apoptosis in h295r adrenocortical cells and in the adrenal cortex of dogs. Endocrinology 157, 1775-1788 (2016).

201. Subramanian, C. et al. Synthetic high-density lipoprotein nanoparticles: a novel therapeutic strategy for adrenocortical carcinomas. Surgery 159, 284-294 (2016).

202. Bussey, K. J. et al. Targeting polo-like kinase 1, a regulator of p53, in the treatment of adrenocortical carcinoma. Clin. Transl. Med. 5, 1 (2016).

203. Martarelli, D., Pompei, P., Baldi, C. & Mazzoni, G. Mebendazole inhibits growth of human adrenocortical carcinoma cell lines implanted in nude mice. Cancer Chemother. Pharmacol. 61, 809-817 (2008).

204. Poli, G. et al. Metformin as a new anti-cancer drug in adrenocortical carcinoma. Oncotarget 7, 49636-49648 (2016).

205. Glover, A. R. et al. MicroRNA-7 as a tumor suppressor and novel therapeutic for adrenocortical carcinoma. Oncotarget 6, 36675-36688 (2015).

206. Babińska, A., Pęksa, R., Wiśniewski, P., Świątkowska-Stodulska, R. & Sworczak, K. Diagnostic and prognostic role of SF1, IGF2, Ki67, p53, adiponectin, and leptin receptors in human adrenal cortical tumors. J. Surg. Oncol. 116, 427-433 (2017).

207. Liu, S., Ding, G., Zhou, Z. & Feng, C. ß-Catenin-driven adrenocortical carcinoma is characterized with immune exclusion. Onco Targets Ther. 11, 2029-2036 (2018).

208. Pennanen, M. et al. C-myc expression in adrenocortical tumours. J. Clin. Pathol. 71, 129-134 (2018).

209. Fernandez-Ranvier, G. G. et al. Identification of biomarkers of adrenocortical carcinoma using genomewide gene expression profiling. Arch. Surg. 143, 841-846 (2008).

210. Aporowicz, M. et al. Minichromosome maintenance proteins MCM-3, MCM-5, MCM-7, and Ki-67 as proliferative markers in adrenocortical tumors. Anticancer Res. 39, 1151-1159 (2019).

211. Cheng, Y., Kou, W., Zhu, D., Yu, X. & Zhu, Y. Future directions in diagnosis, prognosis and disease monitoring of adrenocortical carcinoma: novel non-invasive biomarkers. Front. Endocrinol. 12, 811293 (2021).

212. Mytareli, C. et al. The diagnostic, prognostic and therapeutic role of mirnas in adrenocortical carcinoma: a systematic review. Biomedicines 9, 1501 (2021).

213. Cantini, G. et al. Prognostic and monitoring value of circulating tumor cells in adrenocortical carcinoma: a preliminary monocentric study. Cancers 12, 3176 (2020).

214. Creemers, S. G. et al. Identification of mutations in cell-free circulating tumor DNA in adrenocortical carcinoma: a case series. J. Clin. Endocrinol. Metab. 102, 3611-3615 (2017).

215. Zhang, F. et al. Prognostic role of Ki-67 in adrenocortical carcinoma after primary resection: a retrospective mono-institutional study. Adv. Ther. 36, 2756-2768 (2019).

216. Xu, W. H. et al. Screening and identification of potential prognostic biomarkers in adrenocortical carcinoma. Front. Genet. 10, 821 (2019).

217. Fragoso, M. C. et al. Combined expression of BUB1B, DLGAP5, and PINK1 as predictors of poor outcome in adrenocortical tumors: validation in a Brazilian cohort of adult and pediatric patients. Eur. J. Endocrinol. 166, 61-67 (2012).

218. Sbiera, S. et al. Assessment of VAV2 expression refines prognostic prediction in adrenocortical carcinoma. J. Clin. Endocrinol. Metab. 102, 3491-3498 (2017).

219. Faria, A. M. & Almeida, M. Q. Differences in the molecular mechanisms of adrenocortical tumorigenesis between children and adults. Mol. Cell Endocrinol. 351, 52-57 (2012).

220. Sbiera, S. et al. High diagnostic and prognostic value of steroidogenic factor-1 expression in adrenal tumors. J. Clin. Endocrinol. Metab. 95, E161-E171 (2010).

221. Miller, B. S., Gauger, P. G., Hammer, G. D., Giordano, T. J. & Doherty, G. M. Proposal for modification of the ENSAT staging system for adrenocortical carcinoma using tumor grade. Langenbecks Arch. Surg. 395, 955-961 (2010).

222. Mermejo, L. M. et al. Altered expression of noncanonical Wnt pathway genes in paediatric and adult adrenocortical tumours. Clin. Endocrinol. 81, 503-510 (2014).

223. Lavoie, J. M. et al. Whole-genome and transcriptome analysis of advanced adrenocortical cancer highlights multiple alterations affecting epigenome and DNA repair pathways. Cold Spring Harb. Mol. Case Stud. 8, a006148 (2022).

224. Legendre, C. R. et al. pathway implications of aberrant global methylation in adrenocortical cancer. PLOS ONE 11, e0150629 (2016).

225. Fonseca, A. L. et al. Comprehensive DNA methylation analysis of benign and malignant adrenocortical tumors. Genes Chromosomes Cancer 51, 949-960 (2012).

226. Gao, Z. H. et al. Association of H19 promoter methylation with the expression of H19 and IGF-Il genes in adrenocortical tumors. J. Clin. Endocrinol. Metab. 87, 1170-1176 (2002).

227. Arlt, W. et al. Urine steroid metabolomics as a biomarker tool for detecting malignancy in adrenal tumors. J. Clin. Endocrinol. Metab. 96, 3775-3784 (2011).

Acknowledgements

The authors thank G. Kalafatis for her help with organizing the manuscript draft and for management of the references for the manuscript. The authors apologize to colleagues whose work they may not have cited given the space constraint for the article.

Review article

Author contributions	Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
The authors contributed equally to all aspects of the article.
Competing interests The authors declare no competing interests. Additional information	Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
Peer review information Nature Reviews Cancer thanks Michael Fassnacht, Joakim Crona, who co-reviewed with Liang Zhang, and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.	@ Springer Nature Limited 2023